Stationary walking performance of post-stroke patients and healthy individuals in real and virtual non-immersive environments

Abstract

BACKGROUND:

Several therapies are being used for the rehabilitation of stroke patients, such as Virtual Reality (VR) which has emerged as an interactive intervention to motivate and rehabilitate post-stroke patients. However, data comparison between the virtual and real environments is inconclusive. Thus, this study aimed to compare the kinematics and performance of the affected lower limb of post-stroke patients and healthy individuals during stationary walking activity between the real and virtual non-immersive environments.

METHODS:

A cross-sectional study was conducted with 10 stroke patients and 10 healthy individuals, matched for gender and age. The participants performed stationary walking in a real and non-immersive virtual environment (Wii Fit Plus® –Running mode) for 3 minutes in random order. The performance was measured in both environments using the number of steps, while the kinematics was assessed by calculating the mean maximum flexion and extension of each joint (hip, knee, and ankle) of the affected lower limb.

RESULTS:

Post-stroke patients performed a higher total number of steps (p = 0.042), mainly in the third minute (p = 0.011), less knee flexion (p = 0.001) and total knee range of motion (p = 0.001) in the virtual compared with the real environment.

CONCLUSIONS:

Post-stroke patients performed more steps, with a faster cadence and smaller knee range of motion on the affected side in non-immersive virtual environment compared with the real environment.

Keywords

Cerebrovascular disorders exercise therapy lower extremity psychomotor performance video games

1 Introduction

Stroke is the second leading cause of death and the major cause of disability worldwide. Its incidence has increased vertiginously in high-, middle-, and low-income countries, especially in the elderly population [1, 2].

The most common impairments in stroke patients are balance and gait deficits, which often lead to limitations in activities of daily living and reduced quality of life [3, 4]. In this sense, balance and gait rehabilitation have been proposed to recover functional mobility [5] and movement patterns [6], and improvements depend on the number of sessions and training intensity [7, 8].

It is important to emphasize that there are several approaches to treat the hemiparetic gait and improve motor skills, such as brain stimulation [9] and virtual reality (VR) [10]. Virtual environment systems have been widely used in the last 10 years for the recovery of the affected lower limb after a stroke, presenting satisfactory results in gait, balance and functional mobility [11].

Virtual reality induces neural plastic changes in response to motor areas stimulation, thus recruiting the motor memory system, which contains stored motor programs. Hence, these interactive virtual environment interventions are based on the idea that action processing system stimulation activates cortical areas involved in the execution of movements [12]. Besides, VR provides the therapist options to treat patients in diverse environments, reproducing situations of their routines safely [13]. However, the findings of this technology compared with therapies in real environments are inconclusive, especially concerning gait and balance [14].

One of the justifications for the possible superior effect of VR-based therapy over conventional therapies is credited to motivation, levels of difficulty, and the different environments during exercise, thus increasing the number of repetitions during therapy [15, 16]. Therefore, the knowledge about patients’ performance in virtual and real environments, their differences and similarities, can justify the use of VR-based therapy.

In this context, this study aims to analyze both the kinematics and performance of the affected lower limb during stationary walking between the real and virtual non-immersive environments.

2 Methods

This is a cross-sectional study conducted according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations [17] and the Declaration of Helsinki. The study was approved by the research ethics committee of the institution (number 1.978.573), and all participants signed the informed consent form.

The sample was composed of people aging between 18 and 70 years old, diagnosed by a neurologist with single and unilateral chronic stroke (> 6 months), without auditory, visual, or cognitive deficits, according to the Mini-Mental State examination [18]. Also, the patients should be able to perform functional ambulation with or without supervision, and be classified as level 3, 4, or 5 on the Functional ambulatory Category (FAC) [19].

Healthy individuals paired by age and gender, with no auditory, visual, or cognitive deficits (Mini-Mental State Examination), or musculoskeletal conditions that would affect the gai performance were recruited from the institutional staff. Healthy individuals were included to observe possible differences between environments and individuals who were not undergoing physiotherapy. All participants were recruited between August 2018 and August 2019. The study was conducted in the motricity laboratory of the Faculty of Health Sciences of Trairi (Federal University of Rio Grande do Norte - Brazil).

After meeting the inclusion criteria, the participants filled out a demographic evaluation form to collect data related to age, gender, and time of stroke. Then, the Fugl-Meyer scale was used to evaluate sensory-motor impairment [20], and the Berg Balance Scale (BBS) to assess the static and dynamic balance of post-stroke patients [21].

Finally, performance and kinematic analyses of the affected lower limb were performed in both groups in the virtual and real environments, randomly using sealed envelopes. The activity chosen to assess the lower limb movements in these environments was the stationary walking since it generates the avatar’s gait in many commercially available virtual games. Stationary walking is characterized by carrying out gait movements in the same place, without taking the step forward [22]. Figure 1 shows the study flow diagram.

Fig. 1

Study flow diagram.

The corresponding lower limb of the stroke patient was assessed in the matched healthy individual. Before evaluation, color markers were positioned over the fifth metatarsal head (laterally), lateral malleolus, knee joint line, femur’s greater trochanter, and humerus’ greater tubercle. The marking was done by the same trained researcher.

Stationary walking was performed with a walker positioned in front and a chair behind the patient for safety purposes. A Nintendo Wii® device connected to a 52-inch TV was used to allow the participants to interact with the virtual environment, generating a non-immersive virtual environment.

The Wii Fit Plus® (running mode) game was used for the virtual environment. In this game, the avatar should run in a park-like landscape, meet and overtake other runners along the way. For the avatar to run, the participants should perform a stationary walking with the control attached to their torso. With the device control fastened to the trunk by an elastic bandage, the step movements performed would be captured and reproduced by the avatar on TV (positioned 2.5 meters from the player). The same researcher gave instructions regarding the game and the stationary walking (i.e., walk in the same place without taking a step forward for the avatar to run) to all participants.

The game did not provide any auditory or visual clues of rhythms. The participants had auditory feedback only of the feet touching the floor in the virtual environment. The participant observed himself in third person on the screen, but visualizing the body only from the trunk upwards (not being able to observe the lower limbs of the avatar).

For the real environment, the stationary walking was performed in the same location as the virtual environment; however, the TV was off. The instruction given to the participants was to walk in the same place, without taking the step forward. They were told that the activity should be performed for 3 minutes and that the number of steps would be counted.

Besides, one camera was supported on a tripod, one-meter height, and positioned two meters laterally to the participants. The stationary walking was performed for three minutes in both environments and recorded by video. The recorded videos were imported to the Kinovea software (version 8.20) to evaluate kinematics variables, which presents public domain for the editing and analysis of videos directed to sports. It consists of a two-dimensional kinematic evaluation system [23].

This software was used to draw lines representing the participants’ lower limb segments. The metatarsal and lateral malleolus markers were used to construct the foot segment; the lateral malleolus and knee markers formed the leg segment; the knee and femur’s greater trochanter markers formed the thigh segment; and the greater trochanter and humerus markers formed the trunk segment. Each segment was represented by a line, from one marker to another, while an angle formed by two lines represented a lower limb joint: ankle (foot and leg), knee (leg and thigh), and hip (thigh and trunk lines).

For each joint, the maximum flexion and dorsiflexion angles (swing phase) and the maximum plantar extension and flexion angles (support phase) were evaluated in each step. Then, for each participant was made the mean maximum angles considering the total number of steps. Range of motion was calculated for each joint in the sagittal plane by subtracting the mean maximum extension and mean maximum flexion angles. Figure 2 illustrates the hip, knee, and ankle angles measurement, according to the software used. The performance was measured using the number of steps. All measurements were analyzed in the first, second, and third minutes, and during the total time.

Fig. 2

Hip, knee, and ankle angles representation, according to Kinovea software.

The sample size calculation concerning inter-group analysis was performed according to Miot formula [24]. The mean difference of 26 degrees in knee flexion during the swing phase between post-stroke patients and healthy individuals found by Chen et al. was used for sample size calculation, and the optimal number was estimated as 9 subjects in each group [25].

Data analysis was performed using the Statistical Package for the Social Science software, version 20.0 (IBM Corp.). Initially, demographic and clinic descriptive analysis (age, time of stroke, and score of the clinical scales) were carried out using mean and standard deviation. Subsequently, the Shapiro-Wilk was used to verify the normal data distribution. Comparisons between environments within each group were performed using the paired t-test. Comparisons between groups in each environment were performed using the unpaired t-test. The level of significance was set at 5% (2-tailed) for all statistical tests. The significant results of the comparisons were reported as mean differences with a 95% confidence interval.

3 Results

A sample of 10 post-stroke patients (3 women and 7 men; 6 with left brain injury and 4 with right brain injury) and 10 matched healthy individuals were included (Table 1). None of the participants was excluded or presented symptoms of tiredness and/or pain during and after the exercises.

Table 1
Demographic and clinical data of the participants

Post-stroke patients Healthy individuals

(n = 10) (n = 10)

Age (years) 56,3 (1,05) 59,87 (9,96)

Berg Balance Scale (score) 42,90 (9,30) –

Fulg-Meyer Scale (score) 165,20 (36,48) –

Time of stroke (months) 61,3 (57,00) –

	Post-stroke patients	Healthy individuals
Age (years)	56,3 (1,05)	59,87 (9,96)
Berg Balance Scale (score)	42,90 (9,30)	–
Fulg-Meyer Scale (score)	165,20 (36,48)	–
Time of stroke (months)	61,3 (57,00)	–

Data are shown as mean and standard deviation.

The following variables were statistically different between the real and virtual environments in post-stroke patients: total number of steps (mean difference = –20.10; 95% CI = –39.25, –0.94; t = –2.373; p = 0.042), number of steps 3st minute (mean difference = –8.80; 95% CI = –15.35, –2.24; t = –3.037; p = 0.014), knee flexion (mean difference = –7.69; 95% CI = –11.16, –4.22; t = –5.017; p = 0.001), total knee range of motion (mean difference = 7.88; 95% CI = 4.12, 11.63; t = 4.749; p = 0.001) and total hip range of motion (mean difference = 2.39; 95% CI = .53, 4.25; t = 2.907; p = 0.017). Regarding healthy individuals, the ankle dorsiflexion (mean difference = –1,40; 95% CI = –2.41, –.39; t = –3.138; p = 0.012) was significantly different between the environments.

Besides, inter-group analysis showed statistically significant differences for knee flexion in the virtual (mean difference = 23.44; 95% CI = 5.99, 40.99; t = 2,906; p = 0,012) and real environments (mean difference = 24.26; 95% CI = 4.20, 44.33; t = 2,541; p = 0,020), total knee range of motion in the virtual (mean difference = –29.41; 95% CI = –47.15, –11.67; t = –3,483; p = 0,003) and real environments (mean difference = –28.17; 95% CI = –48,55, –7.79; t = –2,904; p = 0,009), and total hip range of motion in the virtual (mean difference = –11.77; 95% CI = –21.69, –1.86; t = –2,496, p = 0,023) and real environments (mean difference = –12.11; 95% CI = –23.19, –1.04; t = –2,298; p = 0,034) (Table 2).

Table 2

Performance and kinematics during stationary walking movement performed by post-stroke patients and healthy individuals in real and virtual environments

	Post-stroke patients		Healthy individuals
	Real environment	Virtual environment	Real environment	Virtual environment
Number of steps (total)	116.00 (40.81)	136.10 (38.52)^*	138.20 (34.65)	152.40 (34.82)
Number of steps (1st minute)	37.70 (15.52)	42.30 (13.97)	49.50 (12.46)	46.10 (9.76)
Number of steps (2nd minute)	42.00 (13.90)	46.20 (13.77)	47.00 (11.65)	51.00 (11.31)
Number of steps (3rd minute)	36.90 (15.63)	45.70 (11.87)^*	51.50 (11.64)	45.90 (12.50)
Ankle plantar flexion (°)	91.32 (5.18)	91.64 (5.51)	91.32 (5.06)	90.23 (5.86)
Ankle dorsiflexion (°)	86.73 (6.27)	86.90 (5.66)	84.76 (6.56)	86.16 (5.71)^*
Ankle range of motion (°)	4.58 (4.5)	4.73 (3.95)	5.15 (2.70)	5.47 (2.74)
Knee extension (°)	165.78 (9.28)	165.59 (9.16)	169.89 (11.03)	171.76 (9.99)
Knee flexion (°)	136.40 (13.79)⁺	144.09 (10.91)^*⁺	120.64 (23.05)	112.13 (26.86)
Knee range of motion (°)	29.58 (16.90)⁺	21.70 (14.75)^*⁺	57.76 (25.60)	51.11 (22.25)
Hip extension (°)	174.09 (12.12)	173.87 (12.16)	177.28 (7.33)	178.03 (9.08)
Hip flexion (°)	157.55 (12.69)	159.72 (12.66)	149.37 (14.06)	151.35 (11.99)
Hip range of motion (°)	16.54 (9.38)⁺	14.14 (9.38)^*⁺	25.92 (11.60)	28.65 (13.77)

a) Values are presented by mean and standard deviation. b) (°) results shown in degrees. c) ^* Statistical difference between virtual and real environment within each group. d) ⁺ Statistical difference between patients and healthy individuals in each environment.

4 Discussion

This study aimed to identify if there are differences in lower limb kinematics and performance during stationary walking of post-stroke patients and healthy individuals in real and non-immersive virtual environments. The main result found was that post-stroke patients performed more steps, with a faster cadence, mainly in the third minute in virtual environment. However, they also showed a worse movement quality, mainly for knee (smaller maximum flexion in swing phase and a smaller range of motion). The difference found between the environments for the hip range of motion was 2 degrees and would be within bounds of measurement error and, therefore, are not of clinical significance.

It is important to note that the game used for the non-immersive virtual environment did not determine the walking candence. The cadence was determined exclusively by the participant. Also, the system did not offer feedback on performance or results. However, the virtual environment offered visual information from other runners, and the participant could overtake them during the game.

Multisensorial feedback [26] and the increase in people’s engagement levels [27] generated by virtual environments are important mechanisms that may have stimulated post-stroke patients to perform more steps, particularly on the last minute, thus reducing the perception of fatigue. A recent systematic review showed moderate to weak effects of VR on the function of chronic stroke patients under the justification that it is needed at least 8 weeks to obtain a better effect using the VR due to motivation [28]. However, considering the results of this study, it is questioned whether the increased number of repetitions in virtual environment occurs due to a worse quality of movement, manifested by the reduced knee range of motion. Corroborating with our results, Costa et al. observed better performance of post-stroke patients in a virtual darts game (i.e., fewer errors), but with a worse upper limb movement pattern compared with the real environment [29].

Some authors found that the involvement during a virtual experience among post-stroke patients and healthy individuals are similar [30], showing that interaction experience might also be a strategy to improve movement patterns during gait [31]. It is worth mentioning that virtual reality can also help to stimulate improvements in patients’ performance through an increase in cortical excitability and neural regeneration during rehabilitation [32].

Regarding healthy individuals, no statistical significance was found between the virtual and real environments for most analyzed variables. The difference found in dorsiflexion was less than 2 degrees and are not of clinical significance. Many studies have shown that interventions using VR are equally effective than the conventional treatment for individuals without neurological impairments [33].

Considering our results, the use of VR-based therapy through commercially available systems must consider therapeutic goals. If the goal is to increase the number of repetitions, without considering the movement pattern performed in its real context, VR therapy seems to be a good therapeutic tool. The strategy of increasing repetitions is widely used for aerobic training [34]. However, if the goal is to improve the gait movement pattern, the training in a real environment may be more effective. It is interesting to note that the virtual system used in this study is commercially available and was not developed for therapeutic purposes. In this sense, games designed for rehabilitation must encourage the increase of repetitions and the quality of the movement performed.

This study presents some limitations that need to be acknowledged. The evaluation was conducted during a short period (3 minutes). The initial proposal was 5 minutes; however, we conducted a pilot study with two patients and observed dangerous increases in subjective perceived exertion and blood pressure level after 3 minutes. Furthermore, the system used for kinematic evaluation (kinovea) is a 2D system, which does not present the same validity when compared to a 3D system. The findings of this study cannot be interpreted with the level of certainty and rigour than a validated gold standard 3D system. Despite being a system used in scientific research, the reliability of hip and ankle angle measurements presents an error between 2.5° and 5° compared with a 3D system [23]. However, the clinical difference observed in the present study was in the knee joint.

In conclusion, the kinematics and performance of post-stroke patients during stationary walking were different between the real and non-immersive virtual environments. These differences were not found in healthy individuals. Post-stroke patients performed more steps, with a faster cadence and smaller knee range of motion of the affected lower limb in the virtual compared to the real environment.

Conflict of interest

The authors have no conflict of interest to report.

References

Feigin

, Lawes

, Bennett

, Barker-Collo

, Parag

. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol. 2009;8(4):355–69.

Katan

, Luft

. Global Burden of Stroke. Semin Neurol. 2018;38(2):208–11.

Cohen

, Ivanova

, Brouwer

, Miller

, Bryant

, Garland

. Do Performance Measures of Strength, Balance, and Mobility Predict Quality of Life and Community Reintegration After Stroke? Arch Phys Med Rehabil (2018;99(4):713–9.

, Shaughnessy

. The effects of exercise-based rehabilitation on balance and gait for stroke patients: a systematic review. J Neurosci Nurs. 2011;43(6):298–307.

Garland

, Willems

, Ivanova

, Miller

. Recovery of standing balance and functional mobility after stroke. Arch Phys Med Rehabil. 2003;84(12):1753–9.

Kaczmarczyk

, Wit

, Krawczyk

, Zaborski

, Gajewski

. Associations between gait patterns, brain lesion factors and functional recovery in stroke patients. Gait Posture. 2012;35(2):214–7.

Rose

, Nadeau

, Wu

, Tilson

, Dobkin

, Pei

, Duncan

. Locomotor Training and Strength and Balance Exercises for Walking Recovery After Stroke: Response to Number of Training Sessions. Phys Ther. 2017;97(11):1066–74.

Langhorne

, Bernhardt

, Kwakkel

. Stroke rehabilitation. Lancet. 2011;377(9778):1693–702.

Koch

, Bonnì

, Casula

, Iosa

, Paolucci

, Pellicciari

, et al. Effect of Cerebellar Stimulation on Gait and Balance Recovery in Patients With Hemiparetic Stroke: A Randomized Clinical Trial. JAMA Neurol. 2019;76(2):170–8.

10.

Maggio

, Latella

, Maresca

, Sciarrone

, Manuli

, Naro

et al. Virtual Reality and Cognitive Rehabilitation in People With Stroke: An Overview. J Neurosci Nurs. 2019;51(2):101–5.

11.

Chen

, Lo

, Mao

, Ding

, Lin

, Li

, et al. Effect of Virtual Reality on Postural and Balance Control in Patients with Stroke: A Systematic Literature Review. Biomed Res Int. 2016;2016:7309272.

12.

Eng

, Siekierka

, Cameirao

, Zimmerli

, Pykl

, Duff

, et al. Cognitive virtual-reality based stroke rehabilitation. IFMBE Proceedings. 2007;14:2839–43.

13.

Darekar

, McFadyen

, Lamontagne

, Fung

. Efficacy of virtual reality-based intervention on balance and mobility disorders post-stroke: a scoping review. J Neuroeng Rehabil. 2015;12:46.

14.

Laver

, George

, Thomas

, Deutsch

, Crotty

. Virtual reality for stroke rehabilitation (Review). Cochrane Database Syst Rev. 2017.

15.

Saposnik

, Levin

. Virtual reality in stroke rehabilitation: A meta-analysis and implications for clinicians. Stroke. 2011;42(5):1380–6.

16.

Zimmerli

, Jacky

, Lünenburger

, Riener

, Bolliger

. Increasing patient engagement during virtual reality-based motor rehabilitation. Arch Phys Med Rehabil. . 2013;94(9):1737–46.

17.

Von Elm

, Altman

, Egger

, Pocock

, Gotzsche

, Vandenbroucke

. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–9.

18.

Brucki

, Nitrini

, Caramelli

, Bertolucci

, Okamoto

. Sugestões para o uso do Mini Exame do Estado Mental no Brasil. Arq Neuropsiquiatr. 2003;61:777–81.

19.

Mehrholz

, Wagner

, Rutte

, Meissner

, Pohl

. Predictive validity and responsiveness of the Functional Ambulation Category in hemiparetic patients after stroke. Arch Phys Med Rehabil. 2007;88:1314–9.

20.

Fugl-Meyer

, Jääskö

, Leyman

, Olsson

, Steglind

. The post-strokehemiplegic patient. 1. A method for evaluation of physical performance. Scand J Rehabil Med. . 1975;7(1):13–31.

21.

Miyamoto

, Lombardi

Jr , Berg

, Ramos

, Natour

. Brazilian version of the Berg balance scale. Brazilian Journal of Medical and Biological Research. 2004;37(19):1411–4.

22.

Chadefaux

, Moorhead

, Marzaroli

, Marelli

, Marchetti

, Tarabini

. Vibration transmissibility and apparent mass changes from vertical whole-body vibration exposure during stationary and propelled walking. Appl Ergon. 2021;90:103283.

23.

Fernández-González

, Koutsou

, Cuesta-Gómez

, Carratalá-Tejada

, Miangolarra-Page

, Molina-Rueda

. Reliability of Kinovea^® Software and Agreement with a Three-Dimensional Motion System for Gait Analysis in Healthy Subjects. Sensors (Basel). 2020;20(11):3154.

24.

Miot

. Sample size in clinical and experimental trials. J Vasc Bras. 2011;10(4):275–8.

25.

Chen

, Patten

, Kothari

, Zajac

. Gait differences between individuals with post-stroke hemiparesis and non-disabled controls at matched speeds. Gait Posture. 2005;22(1):51–6.

26.

Jang

, You

, Hallett

, Cho

, Park

, Cho

, et al. Cortical reorganization and associated functional motor recovery after virtual reality in patients with chronic stroke: an experimenter-blind preliminary study. Arch Phys Med Rehabil. 2005;86:2218–23.

27.

Lequerica

, Kortte

. Therapeutic engagement: a proposed model of engagement in medical rehabilitation. Am J Phys Med Rehabil. 2010;89:415–22.

28.

Lee

, Park

. The Effects of Virtual Reality Training on Function in Chronic Stroke Patients: A Systematic Review and Meta-Analysis. Biomed Res Int. 2019;2019:7595639.

29.

Costa

, Fernandes

, Oliveira

, Brasileiro

, Ribeiro

, Vieira

, Campos

. Intergame Analysis of Upper Limb Biomechanics of Stroke Patients in Real and Virtual Environment. In Mediterranean Conference on Medical and Biological Engineering and Computing. 2019:610-7.

30.

Borrego

, Latorre

, Alcañiz

, Llorens

. Embodiment and Presence in Virtual Reality After Stroke. A Comparative Study With Healthy Subjects. Front Neurol. 2019;10:1061.

31.

Hasson

, Goodman

. Learning to shape virtual patient locomotor patterns: internal representations adapt to exploit interactive dynamics. J Neurophysiol. 2019;121(1):321–35.

32.

, Ku

, Kim

, Kang

. Virtual Reality-Guided Motor Imagery Increases Corticomotor Excitability in Healthy Volunteers and Stroke Patients. Ann Rehabil Med. 2016;40(3):420–31.

33.

Vogt

, Skjæret-Maroni

, Neuhaus

, Baumeister

. Virtual reality interventions for balance prevention and rehabilitation after musculoskeletal lower limb impairments in young up to middle-aged adults: A comprehensive review on used technology, balance outcome measures and observed effects. Int J Med Inform. 2019;126:46–58.

34.

Lund

, Dalgas

, Grønborg

, Andersen

, Severinsen

, Riemenschneider

et al. Balance and walking performance are improved after resistance and aerobic training in persons with chronic stroke. Disabil Rehabil. 2018;40(20):2408–15.

	Post-stroke patients	Healthy individuals
	(n = 10)	(n = 10)
Age (years)	56,3 (1,05)	59,87 (9,96)
Berg Balance Scale (score)	42,90 (9,30)	–
Fulg-Meyer Scale (score)	165,20 (36,48)	–
Time of stroke (months)	61,3 (57,00)	–