Acute effects of virtual reality exercise bike games on psychophysiological outcomes in college North-African adolescents with cerebral palsy: A randomized clinical trial

Study design

This study was a subgroup analysis of the ongoing RCT registered with the Pan African Clinical Trial Registry (identification number: PACTR202308598603482) on August 31, 2023. The RCT was conducted following the guidelines established by the CONSORT statement.¹² Testing occurred between June and August 2023 at Sfax Association for the Care of the Physically Handicapped (Sfax, Tunisia). All procedures in the study adhered to the ethical standards of the institution and/or national research committee as well as the 1964 Helsinki Declaration and its subsequent amendments or comparable ethical standards.¹³ The study received approval from the University Institutional Review Board and the local ethics committee of the High Institute of Nursing, University of Sfax, Tunisia (Protection Committee Approval in 25 January 2022 Registration code: CPP SUD N 0388/2022). Before commencing the study, adolescents aged 14 to 16 years and their parents or guardians, received an explanation of the test protocol.

Game Selection and Rationale

Although TE bikes have been utilized in numerous rehabilitation settings, the integration of VR technology represents an innovative approach to increasing PA and improving health outcomes in this specific population. The VirZoom VRE system provides a unique and immersive fitness experience that transcends the limitations of conventional stationary cycles by offering users dynamic virtual environments to explore during exercises. This novel approach not only introduces an element of enjoyment and engagement in the training program, but also presents opportunities for cognitive stimulation and social interaction, which are particularly beneficial for adolescents with CP. Furthermore, the VirZoom VRE system enables customizable training protocols and game-based activities to address the diverse needs and abilities of individuals with CP.

The selection of specific games in the VirZoom VRE system was grounded in several rationales backed by literature. Primarily, we aimed to ensure that the chosen games offered a diverse range of PA that engaged various muscle groups and movement patterns, thus addressing the heterogeneous needs and abilities of adolescents with CP. Second, we prioritized games that are immersive, interactive, and engaging, as research has indicated that the entertainment value of VR experiences can significantly influence exercise adherence and participation (e.g., 15). Additionally, we considered the cognitive demands of each game, selecting those that incorporated elements of attention, memory, decision-making, and problem-solving to potentially enhance cognitive outcomes in our participants. Lastly, we aimed to incorporate games that facilitate social interaction and competition, fostering a sense of camaraderie and motivation among participants, which may further contribute to the efficacy of the intervention.¹⁴

Participants

A convenience sample of college students were recruited from the aforementioned association. To select participants, the study’s inclusion criteria were disclosed to physicians, pediatric physiotherapy clinics, special educators, speech therapists, sociologists from the above-cited association, and ambulant care services. Potential participants were approached by association staff and social workers. These professionals recruited them to participate in the study.

The eligibility criteria for participant selection were as follows: i) adolescent aged between 14 and 16 years, ii) medical diagnosis of spastic CP confirmed by a pediatric neurologist; iii) motor function classified at level I or II according to the Gross Motor Function Classification System (GMFCS V), iv) PA levels below the international norm (i.e., less than one hour daily at >5 metabolic equivalents (MET)), indicating moderate or vigorous intensity,¹⁵ v) lack of regular sports participation (i.e., less than three sessions per week for 20 minutes or more), and vi) reported issues with mobility in daily life or sports.

Non-inclusion criteria for participation in the study were: i) engaging in a week of moderate to intense exercise exceeding 150 minutes per week; ii) GMFCS level III to V, iii) behavioral issues that prevent participation in group activities or a movement disorder primarily dyskinetic or atactic in nature, iv) recent surgery within the previous six months, v) botulinum toxin treatment, and vi) Serial casting within the recent three months or scheduled procedures during the intervention period.

Patients’ level of physical ability was categorized using the GMFCS V.¹⁶ GMFCS level I or II indicate independent walking for more than 12 months and the ability to ambulate for at least 10 m with or without a gait-assistance device (walker or crutches). GMFCS level III to V precludes the use of the Oculus Quest handheld controller, moderate to severe intellectual disability, uncontrollable seizures, or conditions that make physical training dangerous (e.g., hip dysplasia, cardiac arrhythmia, or mitochondrial defects).

Randomization

Simple random allocation was used to generate a random allocation sequence. This means that each participant had an equal chance of being assigned to either the VRE or TE group.

No restriction, such as blocking and block size, was applied in this study. This means that participants were not blocked or stratified.

The mechanism used to implement the random allocation sequence (such as sequentially numbered containers) described any steps taken to conceal the sequence until interventions were assigned and sequentially numbered envelopes were used to implement the random allocation sequence. The envelopes were sealed until the intervention was completed. The allocation sequence was generated by the researcher ( MS on the list of authors). The latter also enrolled participants and assigned them to interventions.

Experimental design

The study protocol is illustrated in Figure 1.

Figure 1. Flowchart of the study protocol.

TE: Traditional exercise. VRE: Virtual reality exercise.

Anthropometric measurements, including height, weight, and body mass index (BMI), were recorded for the participants wearing light clothing and no shoes. The participents underwent two 40-minute cycling sessions¹⁷ on two distinct days. A two-day interval separated the sessions, and they were conducted in a counterbalanced manner: (1) a VirZoom VRE bike and (2) a TE bike. Both bikes are Everfit BFK-500 Kaohsiung, Taiwan. Each session consisted of three conditions. The first was a five-minute period in which the participant lay down. This was followed by a 30-minute cycling period. The third condition was another period lying down for the remaining time. The Go-No-Go task⁹ and the Profile Of Mood States scale (POMS)¹⁴ were administered two days prior to each session as well as immediately afterwards. Heart rate (bpm), blood pressure (mmHg), Modified Ashworth Scale,¹⁸ and Montreal Cognitive Assessment (MoCA)¹⁹ scores were recorded immediately before and after each session.

Body composition

Height was measured with a Harpenden 602VR stadiometer to the last complete 0.1 cm, and body composition was estimated using a multifrequency bioelectrical impedance analyzer (TBF-410GS, Tanita Co., Tokyo, Japan). This is a method validated against reference methods.²⁰ The parameters collected in this study included weight, Body Mass Index, fat mass, and fat free mass.

VirZoom VRE

VirZoom’s VZfit offers a unique experience by combining a controller attached to handlebars and a sensor on the bike crank. When used in conjunction with an Oculus Go virtual reality headset, it allows researchers to explore various virtual destinations while the participant is stationary on a bike.²¹

During the study, participants played two 30-minute exercise sessions on the VirZoom VZfit, which features a variety of mini-games. The games chosen for the study were “Le Tour” and “Race Car,” both of which require players to pedal faster or slower to speed up or slow down, and to lean their body side to side to turn left or right. These games were observed to be the most intense and were played with a pedal resistance set at a medium level.

All participants were asked to maintain a moderate level of exertion, with their heart rate ranging between 65% to 85% of their Age-predicted Maximal Heart Rate (ApMHR)²² as measured by a polar sensor monitor.

Traditional stationary exercise bike

To create a training environment similar to traditional cycling and to reduce the environmental impact, participants watched a virtual cycling video on television while using the Sparnod fitness sub-52 stationary bike. VR cycling sessions were designed to achieve an exercise intensity equivalent to 65% to 85% of the ApMHR.²²

To ensure uniform exercise intensity between the traditional and VR cycling sessions, participants were instructed to maintain their heart rate between 65% and 85% of their ApMHR, which was measured by the integrated heart rate monitor on the stationary bike during the traditional cycling session.

The Go/No-Go task

The Go/No-Go task is a response inhibition task in which a motor response must be executed or inhibited.²³ During this task using the Psychology Experiment Building Language (PEBL) computer software, participants button on a keyboard.²⁴ The presentation began with a 2 × 2 array with four stars (one in each square of the array). A single letter (P or R) was then presented in one of the squares for a duration of 500 milliseconds with an inter-stimulus interval of 1,500 milliseconds. In the first condition (P-Go), participants were asked to press a button in response to the target letter P and withhold their response to the non-target letter R. The ratio of targets to non-targets was 80:20. The first condition consisted of 160 trials. A second, reversal condition (R-Go) was then administered, and participants were now asked to make a response to the target letter R and withhold their response to the non-target letter P (the letter that they were initially conditioned to make a motor response to the task, while NoGo errors and RT to Go responses are considered as indicators of impulsivity.²³

The Go/No-Go task is a computerized test that evaluates inhibitory control, a crucial component of cognitive assessment and self-regulation.²⁵ During the task, participants respond by pressing a button when certain visual stimuli appear (Go trials) and withhold their response to other stimuli (No-Go trials), measuring their response inhibition. However, response inhibition can be difficult if No-Go trials are relatively rare, as Go responses become prepotent, meaning the system is biased to produce them. To ensure that inhibitory control is necessary, No-Go trials should make up less than 20% of the trials and each trial’s duration should be shorter than 1,500 ms.²⁶ To interpret data from the Go/No-Go task, precision and reaction time are measured, performance is compared with established norms, and errors of commission and omission are analyzed. Examining these factors provides insight into the participant’s response inhibition and determines whether their performance falls within the expected range.²⁶

The Montreal Cognitive Assessment

The MoCA is a cognitive screening tool designed to detect Mild Cognitive Impairment (MCI) in the early stages of various neurodegenerative disorders.²⁷ It is a 10-minute test that is more sensitive than other commonly used screening tools, such as the Mini-Mental State Examination, making it a reliable and valid tool for detecting MCI in clinical settings.²⁸

The MoCA is composed of 30 questions that assess various cognitive domains, including orientation, executive functioning, memory, language, and visuospatial abilities.¹⁹ We used for this study the Arabic version. The tasks in the MoCA include naming objects, memorizing a short list of words, drawing a clock face, and performing serial subtraction tasks. Scores on the MoCA range from 0 to 30, with a cut-off point of 26, where a score of 26 or higher is generally considered normal.¹⁹

The Profile of Mood State

Mood can be measured using a valid, reliable, and sensitive self-report questionnaire called the POMS,²⁹ which consists of 58 items (i.e., words or sentences) and can provide valuable information about adolescents’ emotional states.³⁰ We used the POMS Arabic version of the POMS.³¹ The POMS is a self-report assessment tool that measures six dimensions of mood: Tension-anxiety, anger-hostility, vigor-activity, fatigue-inertia, depression-dejection and confusion-bewilderment.

The POMS is a 65-item questionnaire that is scored on a 5-point scale (0 = not at all to 4 = extremely).³² It takes about 10-15 minutes to complete.

The Physical Activity Enjoyment Scale (PACES)

The PACES is a reliable, valid and sensitive measure of PA enjoyment across populations.^33,34 The Arabic version of the PACES, which has been reported to be a reliable and valid measure of enjoyment of PA in Arabic-speaking populations, was applied.³⁵ The PACES consists of 18 items that are rated on a 7-point Likert scale, from 1 (I don’t enjoy it at all) to 7 (I enjoy it very much). The total score of the PACES ranges from 18 to 126, with higher scores indicating greater enjoyment of PA.

The Modified Ashworth Scale

The MAS assesses spasticity by measuring resistance to passive joint movement in patients with neurological disorders.¹⁸ The scale assigns scores from 0 to 5 based on the degree of increased muscle tone, ranging from no increase (0) to severe increase (4). To conduct the assessment, participants are placed in a supine position, and for muscles primarily involved in flexion, the joint is moved from maximal flexion to maximal extension over one second, while muscles primarily involved in extension are moved from maximal extension to maximal flexion over one second. The investigator records the level of resistance encountered during the movement to determine the corresponding score: 0 for no resistance, 1 for minimal resistance at the end range, 1+ for a catch followed by less resistance over half the range, 2 for rigidity over more than half the range, and 3 for considerable resistance over most of the range. In this study, we have utilized the 0 to 5 version of the MAS³⁶ to measure spasticity. The latter version enables numerical quantification of spasticity severity. During the assessment, the participants were placed in a supine position, and the investigators slowly flexed and extended each joint over one second, recording the amount of resistance encountered on the scale from 0 to 5.

Sample size and statistical analysis

A priori power analysis was conducted using G*Power3³⁷ with the alpha level set at 0.05 and a power of 0.80, which concluded that the sample size to find significance should be 14 participants.

Values were presented as the mean ± standard deviation (SD). Cohen’s effect sizes (d) were classified as small (0.20), moderate (0.50), and large (0.80).³⁸ A 2 × 2 repeated-measures analysis of variance (ANOVA) was conducted to examine the effects of exercise type (TE vs. VRE) and time (pre- vs. post- exercise) on performance scores. Data were analyzed using SPSS version 28.0. Prior to the analysis, Mauchly’s test of sphericity was performed to assess the assumption of sphericity. If Mauchly’s test was significant, indicating a violation of sphericity, the degrees of freedom were corrected using Greenhouse-Geisser or Huynh-Feldt corrections, as appropriate.³⁹ The main effects and interactions were evaluated, and post-hoc tests were conducted using the Bonferroni correction to further explore the significant effects. Descriptive statistics, including the means and SDs for each condition, were calculated to provide a comprehensive understanding of the data. Statistical significance was set at an alpha level of .05 for all analyses.

Attention

There was a significant main effect of exercise type on attention (F (1,13) = 6.30, p = 0.026, ηp² = 0.327). Participants’ mean response attention was higher in VRE (SD = 1.25137) compared to TE (SD = 1.15073). This difference was a significant main effect of time F (1.13) = 97.066, p = 0.001, ηp² = 0.882). Participants’ mean response attention was higher at post-training (SD = 0.615) compared to pre-training (SD = 1.251). There was a statistically significant interaction between exercise type and time (F (1.13) = 6.30, p = 0.026, ηp² = 0.327).

Memory

The analysis indicated a non-significant main effect of exercise type on memory (F (1,13) = 3.059, p = 0.104, ηp² = 0.190); however, participants’ mean response memory in the TE group (4.14 ± 0.66) was observed to be higher than in the VRE group (3.86 ± 0.86). The main effect of time was significant (F (1,13) = 180.974, p = 0.001, ηp² = 0.933), suggesting a considerable impact of time on participants’ memory scores.

Total score

There was a significant main effect of exercise type on the participants’ total scores (F (1.13) = 19.118, p = 0.001, ηp² = 0.595). Participants’ mean response total score were higher for VRE (SD = 2.547) than for TE (SD = 2.841). This difference was a significant main effect of time F (1.13) = 80.686, p = 0.001, ηp² = 0.861). Descriptive statistics revealed that participants’ mean response attention was higher post-training (SD = 2.547) than pre-training (SD = 2.334). There was a statistically significant interaction between exercise type and time F (1.13) = 19.118, p = 0.001, ηp² = 0.595).

Vigor

The results for vigor indicated no significant effect of exercise type (F (1,13) = 2.576, p = 0.133, ηp² = 0.165), with attention to the absence of mean values obscuring the interpretation of results. Nonetheless, participants reported a higher average vigor post-training regardless of the type of exercise.

Decision-making correct

There was a significant main effect of exercise type on participants’ decision-making (F (1,13) = 26.338, p = 0.01, ηp² = 0.670). Participants’ mean decision-making scores were higher in the VRE condition compared to the TE condition, although means are required for a complete interpretation.

Decision-making errors

Analysis indicated a significant main effect on decision-making errors (F (1,13) = 28.202, p = 0.001, ηp² = 0.684), with errors being fewer in the VRE condition compared to the TE condition. The main effect of time (F (1,13) = 5.350, p = 0.038, ηp² = 0.292) suggested participants made fewer decision-making errors post-training.

MAS

Table 3 shows a comparison of pre- and post- training scores for spasticity in both the TE and VRE training conditions. Both TE and VRE interventions effectively reduced spasticity levels in various lower body regions. Notably, the VRE group exhibited more substantial improvements, with a 5.61 reduction in spasticity for the lower body (ankle and knee) than the TE group (3.28). In the ankle region, VRE resulted in a significant 2.29 reduction in spasticity, whereas TE reduced spasticity by 1.43.

Limitations

Some limitations must be considered. The major one was the retrospective registration of our RCT. Since the primary purpose of trial registration is to enhance transparency, reduce bias, and promote the integrity of clinical research, retrospective registration of an RCT can pose some challenges and may be viewed with skepticism by journal editors, peer reviewers, and readers. Retrospective registration of an RCT can raise concerns about selective reporting, outcome switching, and potential bias. However, we have registered our RCT in the aim to be transparent about the timeline of the study. Registration was performed immediately as possible just before the trial ended. We included the full study protocol with detailed information on the study design, primary and secondary outcomes, and statistical analysis plan. This demonstrate that our study was conducted in a rigorous and pre-planned manner. We confirm that the present study adhered to ethical guidelines and that the retrospective registration does not compromise the ethical conduct of the trial. We explicitly discuss how we have mitigated the risk of bias in our study by addressing concerns related to selective outcome reporting and data-driven decisions. We attest that there were no modifications to the study design or analysis plan after the data collection. Second, the study’s short duration precludes conclusions about prolonged effects. Third, the “small” but calculated sample size may limit generalizability of the results. Finally, potential confounding factors such as participant exercise history and spasticity severity were not considered. Despite these limitations, the present study has enhanced our understanding of the relationship between TEs and VREs in adolescents with CP. We believe that our findings will stimulate further investigations in this important area.

Implications and futures directions

For adolescents with CP, VRE bike games using VirZoom’s VZfit appear to be an effective alternative to boost adherence and, consequently, enhance mood. Future studies should be conducted in this regard to investigate additional topics related to VRE bike games for children, including assessing those with CP using objective measures to determine how long the virtual headset workouts last, assessing the interactions between individual and group activities, and determining the long-term effects of the activities performed both in the lab and at home.

Underlying data

Zenodo: Data of 14 participants in the RCT titled Acute effects of virtual reality exercise bike games on psycho-physiological outcomes in college North-African adolescents with cerebral palsy [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.13952330.⁴⁸

The project contained an Excel file that included the numerical data of the 14 participants.

Reporting guidelines

CONSORT checklist for: Acute effects of virtual reality exercise bike games on psycho-physiological outcomes in college North-African adolescents with cerebral palsy: A randomized clinical trial. Zenodo. https://doi.org/10.5281/zenodo.10157607.⁴⁹

Data are available under the term Creative Commons Zero (CC0).