Skip to main content

Table 4 Outcome measures and findings

From: A systematic review and meta-analysis of the effectiveness of virtual reality as an exercise intervention for individuals with a respiratory condition

Authors, (year) [reference]

Outcome measures

Adverse effect(s)

Limitation

Findings (mean ± SD)

Butler, Lee, Goldstein and Brooks (2019) [6]

HR, energy expenditure, dyspnea, health-related QoL

Not stated

Small number of studies with small sample size

Risk of bias as hard to blind the participants

4 studies used HR as an outcome measure. 3 showed significant improvement after intervention and 1 showed higher MHR compared to CG. 1 study showed no differences in HR comparing VRG to CG

4 studies used energy expenditure as an outcome measure. 1 showed significant improvement after intervention and 1 study showed no differences in energy expenditure comparing VRG to CG

5 studies used dyspnea score as outcome measure. 3 showed significant improvement after intervention and compared to CG. 1 study showed no differences in energy expenditure comparing VRG to CG

2 studies measured health-related QoL as an outcome measure and they all showed improvement after intervention

Carbonera, Vendrusculo, and Donadio (2016) [26]

Primary: HR, Vo2

Secondary: Dyspnea and fatigue, SpO2, and energy expenditure

No adverse event

Small number of studies

Great variance of type of exercise or test used as comparator

Majority of studies used estimation of MHR, only one used objective measurement

No significant between-group difference in HR was found in 75% (n = 3) of included studies that compared HR between group (n = 4)

VRGs achieved % of MHR recommended for training in 75% (n = 4) of the relevant included studies

All VRGs achieved higher energy expenditure than the CGs in the relevant included studies (n = 2, 100%)

Relevant included studies demonstrated a similar (n = 3, 75%) or higher (n = 1, 25%) between group SpO2 measurements

Relevant included studies showed a lower level (n = 2, 50%) or similar level (n = 2, 50%) of dyspnea and fatigue in VRGs

Del Corral, Cebrià I Iranzo, López-de-Uralde-Villanueva, Martínez-Alejos R, Blanco, and Vilaró (2018) [19]

6MWT distance, MSWD, HJT, MBT, HG

Common muscle stiffness

Unsupervised and long follow-up period increases drop-out rate and nonadherence to exercise recommendations

Using field tests instead of laboratory tests as assessments

VRG demonstrated improvement in all outcome measures (effect size: 0.25 to 0.85, p < 0.05)

VRG demonstrated significantly greater improvement in all outcome measures (effect size: 0.99 to 1.96, p < 0.05) than CG

12 months follow up:

VRG showed a better MSWD than pre VRG (effect size: 0.29, p < 0.01) and a greater improvement than follow-up CG (effect size: 0.74, p < 0.05)

VRG showed a better MBT, right HG, and left HG than pre VRG (effect size: 0.54, 1.08, and 0.88 respectively, all p < 0.01)

VRG showed a better right and left HG improvements (effect size: 1.54 and 1.51, p < 0.01) than CG

de Corral, Percegona, Seborga, Rabinovich, and Vilaró (2014) [11]

HR, dyspnea, Fatigue, SpO2,

No adverse event

Lack of an incremental test to use as comparator

Short duration of intervention session

Wii Active and Wii family Trainer VRGs achieved a higher % of predicted MHR (80.1 ± 7.4 and 82.1 ± 7.5 vs 79.8 ± 7.7 bpm, p < 0.01) than 6MWT

No significant differences were found in SpO2 and dyspnea between all VRGs and CG

Wii fit VRG showed a lower fatigue score (1.0 ± 1.3 vs 2.8 ± 2.5, p < 0.01) than CG

Frade, Dos Reis, Basso-Vanelli, Brandão, and Jamami (2019) [24]

Dyspnea, fatigue, SpO2, MHR, VO2 peak, number of steps in STVR

Not stated

Convenience recruitment of sample

Lack of representation of population group

No screening of function impairment which may affect gait

No measurement of participant’s step length

VRG has a higher SpO2 (88.5 vs 85%, p < 0.05) and VO2 peak (13.5 ± 3.3 vs 12.6 ± 3 mL/min/kg, p < 0.05) than CG

No significant differences were found in MHR, dyspnea, and fatigue score between VRG and CG.

Good intra- and inter-rater reliability in VO2 peak (0.80 and 0.57 ICC, p < 0.001) and number of steps in STVR (0.94 and 0.93, p < 0.001)

Gomes, Carvalho, Peixoto-Souza, Teixeira-Carvalho, Mendonça, and Stirbulov (2015) [7]

HR, energy expenditure, treadmill distance and time, lung function

Not stated

Possible underestimation of energy expenditure with the chosen tool

No individualized exercise intensity in the intervention group

VRG showed improvements in all outcome measures (size effect: 0.3 to 1.07, all p < 0.05), as well as CG (except for resting HR)

VRG showed a higher predicted % of MHR than CG (103.2 ± 8.6 vs 96 ± 7.8, p < 0.05)

VRG showed a higher total energy expenditure than CG (159 ± 41.6 vs 133.3 ± 32.1 calories, p < 0.05)

CG showed a higher treadmill distance (895.8 ± 143.4 vs 703.3 ± 148.3 m, p < 0.05) than VRG

Holmes, Wood, Jenkins, Winship, Lunt, and Bostock (2013) [9]

HR, SpO2, dyspnea, and RPE

No adverse event

No objective measures of exercise intensity

Replacing a laboratory treadmill test with a cycle ergometer, i.e. invalid measurement

Small sample size

Exercise with VR showed an 86% of MHR demonstrated in CPET

Less desaturation (p < 0.05) was evident during exercise with VR, comparing to CPET

Lower dyspnea and RPE score (p < 0.05) were evident during exercise with VR, comparing to CPET

Kuys, Hall, Peasey, Wood, Cobb, and Bell (2011) [21]

HR, energy expenditure, SpO2, enjoyment, dyspnea, and fatigue

Not stated

No long-term effect examined

Possible inaccurate measurement of energy expenditure with the armband design of energy expenditure measurement tool

VRG had a higher total energy expenditure (127 ± 55 vs 101 ± 55 kcal, p < 0.05) than CG

VRG and CG showed a similar average HR (144 ± 13 vs 141 ± 15 bpm) during exercise

VRG had a higher enjoyment score (7.3 ± 1.6 vs 4.7 ± 2, p < 0.05) than CG

No significant difference in dyspnea (5.1 ± 2.1 vs 5.1 ± 2.2) and RPE (15.0 ± 2.6 vs 15.5 ± 2.6) were found between VRG and CG

LeGear, LeGear, Preradovic, Wilson, Kirkham, and Camp (2016) [22]

Total energy expenditure, HR, RPE, dyspnea, and SpO2

Not stated

Possible inaccurate measurement of energy expenditure with the armband design of energy expenditure measurement tool

Small sample size

VRG showed a higher SpO2 (94.7 ± 2.5 vs 92.3 ± 3.3%, p < 0.0001) than CG

No significant differences were found in total energy expenditure, HR, RPE, and dyspnea between VRG and CG

Liu, Meijer, Delbressine, Willems, Franssen, and Wouters (2016) [17]

6MWT distance, SpO2, HR, fatigue, and dyspnea

Not stated

CG performed one 6MWT and VRG performed two and the best attempt out of the two was chosen to analyze

Time gap between GRAIL 6MWT and the post HR and SpO2 measurements

Underrepresentation of GOLD stage 4 COPD patients and complex COPD patients in sample

Monocentric study as limited access to the GRAIL

Learning effect of GRAIL 6MWT was not established

Significant differences were found between VGS and CG in all outcome measures in over ground 6MWT (all p < 0.05)

6MWT distance: 511.0 ± 64.6 vs 668.8 ± 73.6 m

Changes in pre- and post-SpO2: −7.1 ± 5.9 vs −1.2 ± 3.4%

Changes in pre- and post-HR: 29.5 ± 11.8 vs 47.3 ± 15.7 bpm

Changes in pre- and post-dyspnea: 4.0 ± 2.3 vs 1.1 ± 0.9 points

Changes in pre- and post-fatigue: 3.7 ± 2.2 vs 1.1 ± 1.0 points

Significant differences were found between VGS and CG in all outcome measures in GRAIL 6MWT (all p < 0.05)

6MWT distance: 483.7 ± 84.5 vs 692.3 ± 62.0 m

Changes in pre- and post-SpO2: −2.0 ± 4.4 vs 0.0 ± 0.9%

Changes in pre- and post-HR: 19.1 ± 10.5 vs 32.6 ± 15.1 bpm

Changes in pre- and post-dyspnea: 3.4 ± 2.2 vs 1.0 ± 0.9 points

Changes in pre- and post-fatigue: 3.2 ± 2.1 vs 1.1 ± 1.0 points

O’Donovan, Greally, Canny, McNally, and Hussey (2014) [25]

MHR, energy expenditure, VO2

No adverse event

Only recruited individuals with cystic fibrosis who were well and had a relatively good lung function

Individuals require oxygen supplement were not recruited due to the requirement of wearing facemask during measurements

No significant differences were found in all outcome measures between VRGs and CG

Salonini, Gambazza, Meneghelli, Tridello, Sanguanini, and Cazzarolli (2015) [23]

HR, SpO2, dyspnea, and fatigue

Not stated

Only one short session of intervention, not enough to prove active gaming provides a sufficient training effect

Less participants in the VRG reached 80% of MHR (40 vs 67%, p < 0.05) than CG

No significant between-group difference was found in SpO2

VRG experiences less fatigue and dyspnea (p ≤ 0.001) than CG

Sutanto, Makhabah, Aphridasari, Doewes, Suradi, and Ambrosino (2019) [20]

6MWT distance, dyspnea, QoL

Stated that all adverse events (including pulse rate higher than the predicted maximum, respiratory rate above 30/min, SpO2 below 90%) were recorded, but did not specify the event

Small sample size, underpowered

No measurement of exercise intensity in VRG

Standard exercise training may have masked the effect of the additional virtual reality gaming exercise

Only the exercise component of traditional pulmonary rehabilitation was included

No blinding applied to participants and assessors

Both VRG and CG demonstrated within-group improvement in 6MWT distance (52.4 ± 20.6, p < 0.0001 and 66.8 ± 27.8, p < 0.0001), and no between-group difference was found

VRG showed a lower dyspnea score (4.5 ± 1.3 vs 5.7 ± 1.3, p < 0.05) than CG at baseline, and no difference was found between group after intervention

Both VRG and CG demonstrated within-group improvement in health-related QoL (27.0 ± 14.3, p < 0.0001 and 24.6 ± 17.3, p < 0.0001), and no between-group difference was found

Simmich, Deacon, and Russell (2019) [13]

HR, SpO2, dyspnea, enjoyment

No studies reported the occurrence of adverse events linked to virtual reality gaming

Small number and significant heterogeneity of included studies

Vulnerable to publication bias as only high-quality studies were included

Possible language bias, only English search terms were used

HKSJ estimation method may produce overestimation

No significant difference was found in HR, dyspnea, and SpO2 between VRGs and CGs after calculation of mean difference

Large effect of enjoyment among VRGs was found comparing to CGs

Sánchez, Salmerón, López, Rubio, Torres, and Valenza (2019) [15]

Lung function, knowledge of condition, QoL, exercise capacity

Not stated

Small number of studies

Heterogeneity of included studies

Knowledge of asthma were significantly improved in all educational VRGs

Increase in exercise capacity, QoL, and improvement of symptoms were found in VRGs

  1. Abbreviations: VRG virtual reality group, CG control group, MHR maximal heart rate, HR heart rate, RPE rating of perceived exertion, 6MWT 6-min walking test, SpO2 oxygen saturation, kcal kilocalories, bpm beat per minute, HJT horizontal jump test, MBT medicine ball throw, HG hand grip, MSWD modified shuttle walk test distance, STVR stationary walk test with virtual reality, GOLD Global Initiative for Chronic Obstructive Lung Disease, QoL quality of life, CPET cardiopulmonary exercise test, VO2 oxygen consumption