A combination of multimodal physical exercises in real and virtual environments for individuals after chronic stroke: study protocol for a randomized controlled trial

This is a prospective, randomized, single-blind trial of a 15-week exercise program to investigate the effects of the combination of physical exercise in real and virtual environments for chronic post-stroke individuals. It was approved by the Ethics and Research Committee on Human Beings of the School of Physical Education and Sports at the University of São Paulo (CAE no. 40,688,114.9.0000.5391) and is registered with ensaiosclinicos.​gov (no. RBR-4pt72m).

All participants and/or their relatives will provide written informed consent prior to participant enrollment. A Consolidated Standards of Reporting Trials (CONSORT) flow diagram of the trial is shown in Fig. 1, and a Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist is provided in Additional file 1.

A prospective, randomized trial with concealed allocation, blinded assessors, and intention-to-treat analysis will be carried out. The study will be conducted in the Motor Behavior Laboratory in the School of Physical Education and Sports at the University of Sao Paulo, Brazil.

Participants will be recruited from the waiting list for the rehabilitation program at the Physical Education and Sports School in Sao Paulo, Brazil, and will be screened for eligibility by neurological rehabilitation physiotherapy specialists.

Screening will be done to obtain personal data for the individuals (name, sex, date of birth, address, telephone, and schooling), medical history, brain injury topography, and type of locomotion. After this initial screening, the individuals will be referred for a specific medical evaluation in order to investigate their health conditions, request the necessary tests, and follow the medicines used. Only those individuals who receive medical release for the practice of physical exercises will enter the study.

The inclusion criteria will be chronic phase stroke (more than 6 months post-stroke, according to Bernhardt et al. [25]), territory of the lesion in the middle cerebral artery or anterior cerebral artery, both types of stroke (ischemic and hemorrhagic), cognition greater than 24 points on the Mini Mental State Examination, at least 2 months of noninvolvement in other structured forms of physical exercise intervention (community), and no experience in virtual reality games.

The exclusion criteria will be individuals with any type of cardiovascular complication that would contraindicate physical exercise, individuals who underwent surgeries to attenuate clinical conditions resulting from stroke, and those who have undergone chemical blockade to reduce spasticity.

Individuals who fit the study requirements will receive the informed consent form, which will be duly read and explained by the researcher and signed by the individuals and/or their legal guardians.

Participants who meet the study inclusion criteria will be randomized into one of three sample groups—the real multimodal group (RMG), the virtual multimodal group (VMG), or the combined multimodal group (CMG)—through a draw of numbered (generated by computer) and sealed opaque envelopes. The evaluators have no knowledge of the results of the individual allocation (blinded evaluators). However, the same is impossible to perform with the professionals who will perform the interventions; because the interventions are supervised and progressive, there are clearly distinct protocols between groups of individuals.

To characterize the individuals, the Mini Mental State Examination, Fugl-Meyer Assessment, and Orpington Prognostic Scale will be applied. The individuals will also be characterized by the demographic variables of sex, age, type of stroke, injury time, affected brain hemisphere, and schooling.

The primary outcome will be the domain “activities of daily living” from the Stroke Impact Scale (SIS). This will be used to measure the individuals’ levels of participation. As secondary outcome measures, the other domains from the SIS will be used, as well as their total score and the percentage of recovery from stroke, for the purpose of measuring the range of structures/functions and activities. These will be measured using the Montreal Cognitive Assessment (MoCA), Berg Balance Scale (BBS), 6-minute walk test (6MWT), 10-meter walk test (10MWT), and Timed Up & Go Test (TUG).

All measurements will be carried out by trained evaluators who will not be involved in the interventions and will be blinded to the allocation of individuals to the groups. All individuals will be evaluated at three moments: before the beginning of the interventions (baseline), immediately after the end of the interventions (posttest), and 1 month after the end of the interventions (follow-up). The schedule of enrollment, allocation, and postallocation is provided in Table 1.

The intervention for the three sample groups will be 15 weeks long and will consist of two weekly sessions of 60 min each. All interventions will be carried out at the School of Physical Education and Sports of the University of Sao Paulo, Brazil.

The individuals allocated to RMG will carry out sessions in multimodal training in a real environment. The sessions will be divided into four moments, based on guidelines from Billinger et al. [26]: balance and cognition (15 min in duration), training of the aerobic component and muscular strength (20 min each), and a fourth moment focused on flexibility and relaxation (5 min in duration). Aerobic circuits, games, adapted sports, strength training stations, tasks with varying bases of support, double tasks, memorization tasks, and other strategies will be used. Materials will include balls of diverse sizes and constitutions, dumbbells, washers, cones, hula hoops, bladders, wooden sticks, mats, shuttlecocks, and rackets.

Individuals assigned to the VMG group will carry out individualized multimodal training sessions in a virtual environment. For this protocol, there will be eight games of Stability and Balance Learning Environment, a system of virtual reality developed by Motekforce Medical (Amsterdam, the Netherlands), which is composed of a 10.76-ft² force platform, three projectors, six infrared cameras, a sound system, and a touchscreen panel for selection and control of variables. The games of this virtual reality device were produced and adapted to be beneficial for special populations that have balance and movement disorders, such as people with neurological disorders (stroke, cerebral palsy, Parkinson’s disease, traumatic brain injury), orthopedic disorders (amputees, osteoarthritis, musculoskeletal disorders), and the geriatric population (elderly with increased risk of falling). Stability and Balance Learning Environment requires the player to move the entire body, which includes tasks of reach, stationary gait, and others.

These games were grouped into two blocks of sessions that will be applied alternately throughout the intervention. The odd sessions (Fig. 2) will include the games balloon pop, city ride, hit the mole, and 2d maze, and the even sessions (Fig. 3) will consist of the games road encounters, road stepping, paper flight, and hit knees. The number of selected games as the group session in two models aims to make the most motivating interventions possible and keep the characteristics between the games of both models similar.

In order to guarantee the multimodal character of the virtual intervention and to make it similar to the real intervention, the games were classified according to the perceptual and motor demands of the virtual reality systems protocol developed by Cairolli et al. [27]. The use of this classification protocol ensured that the games placed similar demands on the individuals. For example, the demand for strength and cardiovascular resistance could be consistent across the selected games.

Finally, individuals in the CMG group will perform an intervention that combines the protocols of the two other groups described above. In this way, one of the weekly sessions will be held in a real environment and the other in a virtual environment.

Before the start of all sessions, regardless of the group, the individuals’ blood pressure will be verified in order to ensure that no one initiates the practice of physical exercise with a blood pressure greater than 160/105 mmHg [28]. This contributes to the safety of the practice. To register and control the data, stethoscopes (Efficacy line, Bunzl Saúde, Brazil) and manual blood pressure cuff aneroid sphygmomanometers (Solidor, Bunzl Saúde, Brazil) will be used.

Aerobic exercise intensity control will be performed to ensure the safety of the interventions in the three groups. During the interventions, the individuals’ heart rate (HR) will be monitored constantly in order to keep it aligned with the range prescribed individually using the Karvonen formula ((HR_maximum − HR_rest) × percentage HR intensity + HR_rest) [29]. We will use 40% and 70% of the reserve HR, as determined by Billinger et al. [26], as the lower and upper intensity thresholds, respectively. The maximum HR will be calculated using the formula 220 − age in years [30], except in individuals who use beta-blockers. Thus, the maximum HR will be calculated by the formula 164 − 0.7 × age in years [31]. For the control of HR ranges, we will use Polar brand H7 frequency meters synchronized with the Polar Team application (Polar, Bethpage, NY, USA). As a complementary form to intensity control, we will use the Borg Rating of Perceived Exertion [32, 33], which correlates with the objective measures of the workload level and HR [24]. The individuals’ perception of effort will be verified before the beginning of each session and soon after the onset of the aerobic component in the real environments, or soon after the practice of each of the games in the virtual environment sessions. We will adjust the offer of stimuli that indicate perceptions between values 11 and 14, as determined by Billinger et al. [26].

Resistance exercise intensity will also be controlled according to Billinger et al. [26]. All individuals will perform eight to ten exercises, preferably involving the major muscle groups. Each exercise will include 1 to 3 sets of 10 to 15 repetitions.

Every 5 weeks, the aerobic and resistance exercise intensity will be adjusted in terms of progression. By the end of the 15-week intervention period, we will have performed two progressions. The aerobic component will include gradual increases (10%) of the HR percentage, and the resistance component will increase through larger numbers of sets or repetitions for each exercise.

Quality of life will be evaluated through the SIS. This scale evaluates quality of life by measuring 59 items categorized in 8 domains: strength (4 items), hand function (5 items), activities of daily living/instrumental activity of daily living (10 items), mobility (9 items), communication items [7], emotion (9 items), memory and reasoning (7 items), and participation function (8 items). Each item is rated on a 5-point Likert scale in terms of difficulty the patient has experienced in completing each item. Summative scores will be generated for each domain, ranging from 0 to 100. An extra question will be asked in order to know, on a scale from 0 to 100, how much the individual feels they have recovered since their stroke. Our primary outcome measure will be the domain of the activities of daily living/instrumental activity of daily living, and the rest of the domains will be used as secondary outcome measures [34]. The Brazilian version of SIS 3.0 has satisfactory internal consistency, test-retest reliability, convergent validity, and discriminant validity in stroke patients [34]. SIS 3.0 is a specific measure of psychometrically robust HRQoL that can be useful in assessing the consequences of stroke in different cultural contexts [34]. For some of the domains that make up this scale, there are values of minimum detectable change (MDC) and clinically important difference (CID). According to Lin et al. [35], the values are strength (MDC 24.0 and CID 9.2), activities of daily living/instrumental activities of daily living (MDC 17.3 and CID 5.9), mobility (MDC 15.1 and CID 4.5), and hand function (MDC 25.9 and CID17.8).

Cognition will be evaluated through the MoCA. MoCA was developed as an instrument of rapid screening for mild cognitive dysfunction and for evaluating different cognitive domains, such as attention and concentration, executive functions, memory, language, constructional visual skills, conceptual thinking, calculations, and logical reasoning. The scale ranges from 0 to 30 [36]. In the chronic phase of stroke, MoCA has a good correlation with other short cognitive tests and shows high sensitivity and specificity in the prediction of post-stroke cognitive deterioration [37]. There is no consensus on the cutoff to define cognitive impairment for stroke, but most authors use a 26-point cutoff. There are no MDC values for this scale [38]. For the subacute phase of stroke, the scale shows excellent internal consistency (α = 0.78) [39].

Balance will be evaluated through the BBS, which organizes the quantitative description (classification of 0 to 4) of functional balance ability into 14 items common to daily life. It has a minimum score of 0 and a maximum of 56 [40]. The measurement’s MDC is 4.66 [41], and it has excellent test-retest reliability (intraclass correlation [ICC] = 0.98) [42].

Gait functional capacity will be evaluated through the 6MWT, which quantifies the maximum distance, expressed in meters, that the individual is able to walk in 6 min [43]. The measurement’s MDC is 36.6 m [44]; the CID is 34.4 m [45]; and it has excellent test-retest reliability (ICC = 0.99) [44].

Self-selected gait speed will be evaluated through the 10MWT, which determines the average speed, expressed in meters per second, that the individual applies for a distance of 10 m. The measurement CID is 0.06 m/s [46], and it has excellent test-retest reliability (ICC = 0.94) [44].

Mobility will be evaluated through the TUG, which determines the time, expressed in seconds, that the individual takes to perform a task involving a change of direction, transfer, and gait [47]. The measurement’s MDC is 2.9 s, and it has excellent test-retest reliability (ICC = 0.96) [44].

It was estimated that a sample of 36 participants (12 per group) would provide 88% power (α = 5%) to detect a difference between group means of 17.3 (MDC of the activities of daily living/instrumental activities of daily living domain of the SIS) using the program G*Power 3 [48], which takes into account the number of sample groups and the number of evaluation measures. To reach this total number of individuals, at least 43 individuals must be recruited, allowing a dropout rate of 20%.