A single-blinded randomized controlled trial (RCT) was conducted (NCT04171908) following the CONsolidated Standards of Reporting Trials (CONSORT) guidelines. Non-probabilistic sampling of consecutive cases was used. The sample was randomized, after using QuickCalcs GraphPad Software, into a control group (CG) who received conventional rehabilitation treatment and an experimental group (EG) who received VR treatment with LMC in addition to their conventional rehabilitation sessions. All interventions were performed at the Leganés Association of Multiple Sclerosis and Toledo Association of Multiple Sclerosis in Madrid and Toledo (Spain).
The study inclusion criteria were as follows: a diagnosis of MS according to the McDonald criteria  with over two years evolution; a score of between 3.5 (moderate incapacity, although totally ambulant, self-sufficient, and active for 12 hours/day) and 7.5 (unable to take more than a few steps. Restricted to wheelchair and may need aid in transferring. Can wheel self but cannot carry on in standard wheelchair for a full day and may require a motorized wheelchair) on the Kurtzke Expanded Disability Status Scale (EDSS); with stable medical treatment during at least the six months prior to the intervention; muscle tone in the upper limbs not greater than two points on the modified Ashworth Scale (moderate hypertonia, increased muscle tone through most of the range of movement, but affected part easily moved); as well as a score of four points or less in the “Pyramidal Function” section of the EDSS functional scale; absence of cognitive decline; with the ability to understand instructions and obtaining a score of 24 or more in the Mini-Mental Test; and a score of two points or less in the “Mental Functions” section of the EDSS.
The exclusion criteria were a diagnosis of another neurological illness or musculoskeletal disorder different to MS; the diagnosis of a cardiovascular, respiratory, or metabolic illness or other conditions which may interfere with the study; suffering a flare-up or hospitalization in the last three months prior to commencement of the assessment protocol or during the process of the therapeutic intervention; receiving a cycle of steroids, either intravenously or orally, six months prior to the commencement of the assessment protocol and within the study period of intervention; receiving treatment with botulinum toxin in the six months prior to the beginning of the study; or the presence of visual disorders noncorrected by optical devices.
This protocol was approved by the local ethics committee of the Rey Juan Carlos University. Informed consent was obtained from all participants included in this study.
All groups received the intervention between November and January of 2019. Both the EG and the CG received two 60 min sessions per week over a ten-week period (a total of 20 sessions for each group).
CG received a specific UL intervention by two physical therapists based on conventional motor rehabilitation therapy (60 minutes) based on shoulder, elbow, wrist, and finger mobilization, strengthening of UL extensor muscles and stretching exercises for UL flexor muscles [7, 8], and with functional task practice trying to imitate the movements of the serious games designed for the experimental group (i.e. reaching movements, dexterity, grasping and pincer grasp movements using objects of daily living, such as coins, keys, balls, cups, plates) [10,11].
The EG received the same conventional motor rehabilitation therapy (45 minutes) plus LMC (15 minutes) by two physical therapists while seated at a table placed at mid-trunk height and with the elbow placed at an initial 90° elbow flexion. When necessary, manual assistance by the physical therapist was provided.
All interventions took into account the degree of fatigue experienced by the patients introducing rest time periods.
The video games presented in this paper were developed using Unity3D Game Engine software. A small USB peripheral device that supports hand and finger movements as an input denoted as the Leap Motion Controller (LMC) was chosen as the hand motion capture sensor. This sensor offers excellent precision when tracking different parts of the hand, including movements and positions of the joints of the fingers and the palm of the hand. It must be highlighted that the LMC does not require hand contact or touching for interaction.
Description of the video games
For this study, six serious games were developed by the UC3M authors, according to the guidelines provided by clinicians. The development of each video game aimed to imitate exercises and movements commonly included in conventional rehabilitation, such as palmar prehension, finger flexion and extension, or hand pronation-supination Additionally, some cognitive load when training was included through memory exercises. The Leap Motion sensor was employed to capture the user’s hand movements, and different virtual environments were created using Unity3D Game Engine software. In total, six video games were developed. The games were performed firstly unilaterally (each hand separately) and then bilaterally (both hands at the same time). Figure 1 presents the whole set of video games used in this protocol: the Piano Game (PI), the Reach Game (RG), the Sequence Game (SG), the Grasp Game (GG), the Pinch Game (PG), and the Flip Game (FG). A full description of these games is provided in a previous study . However, the main features of each game are described as follows:
PI: This serious game represents a virtual piano keyboard with ten keys, each corresponding to a single finger on each hand (see Figure 2a). The user must play the piano key that is illuminated with the corresponding finger. During the game, the keys light up first in an orderly sequence, from the little finger to the thumb, and then in a random sequence. For each key pressed correctly, one point is added to the total score. Higher scores equate to better game performance.
RG: This serious game encourages the user to reach for several cubes shown in different spatial positions, placed within the reaching range of the user’s upper extremity (see Figure 2b). A highlighted cube indicates the target to be touched. When the user reaches the cube, it falls to the floor of the virtual scene. The cubes are randomly highlighted after a target is reached. The cubes on the screen are placed at different heights and depths within the user’s workspace. To complete the game, the user must reach all cubes.
SG: This serious game uses the same scenario as the RG game. The user observes a certain sequence of cubes, which is then reproduced through a color change to the cubes that appear on the screen; the user must memorize the sequence and subsequently reach for the cubes in the same order. Furthermore, this game includes exercising visual sequential memory as well as the physical skills that are trained by the RG.
GG: This game encourages the patient to perform movements of closing and opening of the hand (grasping) in coordination with reaching movements. For that purpose, a set of cubes are arranged in a specific pattern, including a red circle in the central part of the screen (see Figure 2c). When a cube is highlighted, the user must grasp the cube and move it to the red circle while keeping their fist closed. Once the cube and the red circle come into contact, the user must open the hand with all the fingers stretched to release the cube. The cube may only be released when it touches the red circle.
PG: This game was designed to improve bidigital grip through the performance of a pinching movement between the thumb and the index fingers. As in the previous games, a cube highlighted is presented in the center of the screen and the user must make the cube smaller, using a pinching movement, until the cube disappears (see Figure 2d).
FG: In this video game, the user must place the palm of the hand over the Leap Motion device imitating a waiter holding out a tray (Figure 2e). A small tray with a cube appears in the center of the screen. The patient must turn the palm downwards. Doing this tray rotation, the cube detaches from tray and it falls to the bottom of the screen.
Altogether, the use of these video games aims to be as non-exclusionary as possible. The games are easy to customize according to patient and rehabilitation needs. For that purpose, a setting menu is included in each videogame in order to set the parameters to best fit the user’s capabilities or limitations. The settings can be defined by therapists at the beginning of the training session, or during the performance of the video game.
All assessments were performed by three physical therapists trained in the use of the measures and blinded to the intervention received by the subjects. The following outcome measures were used in both groups, both at the beginning, at the end of the intervention and in a follow-up period of one month without receiving any treatment for both groups.
Grip strength. A Jamar® hydraulic hand dynamometer was used to measure grip strength. This dynamometer offers accurate and repeatable grip strength readings scaled in pounds and kilograms. All the patients performed three grip movements, and the mean values were recorded. The data for the less and more affected sides were recorded in kilograms. The Jamar® hydraulic hand dynamometer is one of the most commonly used objective tools to assess grip strength, and is considered a device of excellent reliability, sensitivity, and ease of use. It is recommended by the American Society of Hand Therapists and by the Brazilian Society of Hand Therapists .
The Box and Blocks Test (BBT) was performed to measure unilateral gross manual dexterity on both the less and more affected side. The BBT consists of moving the maximum number of blocks from one compartment of a box to another, one by one, within one minute. The BBT is a quick, simple, and reliable measurement of manual dexterity. Its administration procedure is standardized, and its validity has been shown in elderly subjects with upper limb disability [13,14].
The Purdue Pegboard Test (PPT) was used for the assessment of fine manual dexterity, gross dexterity, and coordination. This test evaluates the speed and motor dexterity of each hand and the manual dexterity using both hands at the same time. The PPT features a board with two columns with 25 holes each and a specific number of pins, washers, and collars placed in four containers across the top of the board. The test consists of inserting as many pins as possible in three distinct phases, with a time limit of 30 seconds for each part. First, the test is performed with the dominant hand, then with the non-dominant hand, and then with both hands at the same time. The number of pins inserted is recorded [15,16].
Nine Hole Peg Test (NHPT) was used. It is a hand function test, which consists of a plastic peg board (25.0 cm × 12.7 cm × 2.3 cm) with nine holes (2.54 cm between the holes) and nine pegs (3.2 cm long, 0.64 cm wide). The participant has to put the nine pegs in the peg board as fast as possible, one at a time with one hand only, and then remove them again. The test is performed two times per hand, with the non-affected hand first. The time it takes to fulfill the second trial with the more-affected hand is used for the analysis .
Fatigue Severity Scale (FSS). The FSS described by Krupp et al,  is one of the most commonly used scales for the assessment of fatigue in MS attributed to a multifactorial origin. It consists of nine items that are assessed by the patient with a score between 0 and 7. The cut-off point of this scale is arbitrary, with a score of 5 used by most authors as the reference value to distinguish the presence or absence of the symptom. The result is interpreted as a percentage measure.
Multiple Sclerosis Impact Scale (MSIS-29). This scale is a specific instrument that allows for assessing the physical and psychological well-being of subjects with MS. It is made up of 29 questions divided into two components: a physical magnitude comprising the first 20 questions, and a psychological magnitude with the last 9 questions. The answers are scored on a Likert scale from 1 to 5, with a maximum of 100 points in the physical part and 45 points in the psychological evaluation . The results are interpreted as a percentage measure. The MSIS-29 has demonstrated its validity and suitability for the evaluation of people with MS, compared to other established measures . It is considered a reliable method doe assessing quality of life within the field of MS . The MSIS-29 scale is within the 20 specific scales validated for the measurement of quality of life in the context of MS and is among the three most commonly used according to a number of articles published in this regard .
Satisfaction and adherence. The Client Satisfaction Questionnaire (CSQ-8) was used to evaluate the satisfaction of health service users for both groups. This is a self-administered post-treatment questionnaire, comprised of eight items that evaluate the level of satisfaction regarding the care and quality of the service received and the level of fulfillment of the patient’s expectations regarding the treatment administered. The total score of the questionnaire is 32 points, with higher values meaning higher satisfaction with the treatment received [22,23]. The result is calculated as a percentage measure. In addition, the EG completed a satisfaction questionnaire experimental related to the LMC treatment program. It was designed by the research group based on previous publications on using video games in MS . The questionnaire was composed of 18 items that assess the degree of satisfaction in the following dimensions: technical quality and operation of the equipment (4 items); ease of the video game to be played even in disadvantageous conditions (5 items); program compliance and compliance, and compliance in relation to the treatment performed and its applicability (7 items); general degree of satisfaction or complacency (2 items). The answers of this questionnaire are established on a five-point Likert scale, from not satisfied (= 1) to very satisfied (= 5), with alternative directionality to reduce stereotyped responses. Regarding the interpretation of the results of the surveys, the total score was calculated as a percentage measure.
Additionally, we recorded the attendance rate (%) for therapy sessions (compliance).
The statistical analysis was performed using the SPSS statistical software system (SPSS Inc., Chicago, IL; version 22.0). The Shapiro-Wilk test and the Kolmogorov-Smirnov test were used to screen all data for normality of distribution. The Friedman test was used, which is a non-parametric test for repeated measurements in related samples. In the event that there were significant differences, the Wilcoxon test was performed, which allows two related samples to be compared. Additionally, the Mann-Whitney test for non-related samples was used to compare variables. The statistical analysis was performed with a 95% confidence level, and significant values were considered as p <0.05.