A double blinded randomized controlled trial (RCT) (NCT05895734) was conducted according to the recommendations of the CONSORT guidelines [16]. Participants were randomly assigned to either the control group or the experimental group by a blinded investigator who was not involved in the intervention. The randomization process was performed using the Research Randomizer Version 4.0 program. The control group received conventional physiotherapy treatment, while the experimental group received the same conventional physiotherapy treatment plus treatment with the Powerball® system. All outcome measures in both groups were evaluated by blinded assessors to the intervention.

The present study received ethical approval from the Human Ethics Committee of the Rey Juan Carlos University.

All subjects in the present study were recruited from two multiple sclerosis patient associations in the Community of Madrid (ALEM and AFEM, Madrid, Spain). Participants in this study fulfilled the following inclusion criteria: between 20 and 70 years of age, diagnosed with multiple sclerosis according to the McDonald criteria [17] with a disease duration of more than two years; EDSS score between 2 (minimal disability in one functional system) and 7 (inability to walk more than 5 m, even with assistance, limited to wheelchair mobility, although able to propel oneself and transfer without assistance; active in wheelchair for at least 12 h per day); stable medical treatment for at least six months prior to the intervention; upper limb muscle tone not greater than 2 points (moderate hypertonia, increased muscle tone during the majority of the range of motion but can be moved passively with ease in the affected limb) on the modified Ashworth scale [18, 19]; upper limb muscle balance equal or greater than 3; a score of 4 or less on the “Pyramidal Function” section of the EDSS functional scale; no cognitive impairment; ability to understand instructions and score of 4 or greater on the Mini-Mental State Examination; and a score of 2 or less on the “Mental Functions” section of the EDSS.

Exclusion criteria: diagnosis of any neurological disease or musculoskeletal disease other than multiple sclerosis; diagnosis of any cardiovascular, respiratory, or metabolic disease or other condition that might interfere with this study; history of exacerbation or hospitalization within the three months prior to the start of the assessment protocol or during the therapeutic intervention process; history of botulinum toxin treatment within the six months prior to the start of the study; presence of uncorrected visual impairment.

All patients voluntarily signed an informed consent form to participate in the study.

Participants in the control group received conventional physiotherapy treatment, while participants in the experimental group received conventional physiotherapy treatment plus treatment using the PowerBall® system. Both groups received 45 min of intervention per session and the same total number of sessions.

The conventional physiotherapy treatment consisted of a total of 16 sessions, with 2 sessions per week, over a period of 8 weeks. The treatment consisted of the following components [20,21,22]: osteokinematic mobilizations of the shoulder, elbow, and wrist joints; training activities targeting upper limb manipulative and functional skills for activities of daily living; an upper limb strengthening program, followed by a stretching program for the involved muscles. Each session lasted 45 min.

The experimental group received intervention with the PowerBall® 250 Hz gyroscope system in addition to conventional physiotherapy. The intervention consisted of 3 sets of 1-minute duration with each hand during the first 4 weeks, and 5 sets with each hand in subsequent sessions until the end of the study. The intervention always started with the dominant hand or the less affected hand (Fig. 1).

Protocol for intervention with the Powerball® system

LAS Less Affected Side; MAS More Affected Side

* In each of the sessions, the exercises are performed first with the less affected side and then with the more affected side

As there are different models of PowerBall® and forms to initiate the gyroscope system, the manual activation model with a frequency of 250 Hz was used in this study (Fig. 2).

During the sessions with the PowerBall®, the physiotherapist was responsible for activating the device and initiating the rotation of the system. During all sessions with the device, the patient was seated upright in a chair with a backrest and no armrests to avoid interfering with the position of the working upper limb. The exercise was performed with both upper limbs, alternating between the more affected and less affected upper limb (the more/less affected side was determined by their neurologist based in their EDSS and physical exploration). The rotation of gyroscope was clockwise for the right upper limb and counterclockwise for the left upper limb. The upper limb performing the exercise was supported on a table, with the elbow flexed at 90 degrees, the shoulder flexed at 20 degrees, and the shoulder abducted at 30 degrees. The wrist was in slight extension, with forearm pronation and slight radial deviation, maintaining a comfortable position throughout the exercise. The non-exercising upper limb was placed in the patient’s lap.

Prior to the start of the sessions, a demonstration session of the device was given to familiarize the patients with the PowerBall® system.

All interventions considered the level of fatigue experienced by the patient, allowing for longer rest intervals if fatigue exceeded a score of 7 on the Modified Borg Scale [22].

The following assessment tools were used at the beginning and end of the intervention in both groups by researchers who were not involved in the intervention. Patients were then reassessed 3 weeks after stopping Powerball® but continuing with conventional physiotherapy. Isometric Grip Strength: The JAMAR® hand dynamometer was used to measure grip strength. The JAMAR® dynamometer is widely recognized as the gold standard for grip strength assessment [23, 24]. It has demonstrated excellent test-retest and inter-rater reliability across different populations [24, 25]. The patient held the dynamometer with the upper limb in 0° abduction and rotation, the elbow flexed at 90°, and the forearm in a neutral position. The wrist was held between 0 and 15° of radial deviation, and a maximum grip was exerted for 3 s using all five fingers. The force exerted was measured in kilograms [25]. Three measurements were taken with each hand, always starting with the dominant hand, and the average of the three measurements was calculated [25].

Isometric Pinch Strength: A Baseline Pinch Gauge® was used to assess three different types of pinch grip. The two-finger opposition grip [26] in which the anterior and distal surfaces of the first finger contact the anterior surface of the pinch gauge and the anterior and distal surfaces of the second finger contact the posterior surface of the pinch gauge; the lateral pinch grip [27] in which the anterior surface of the distal phalanx of the first finger contacts the anterior surface of the pinch gauge, and the radial edge of the second finger contacts the posterior surface of the pinch gauge; and three-finger pinch grip [28] in which the anterior and distal surfaces of the first finger contact the posterior surface of the pinch gauge, and the anterior and distal surfaces of the second and third fingers contact the anterior surface of the pinch gauge. Three maximum force pinches were performed for 3 s each, alternating between the dominant and non-dominant hand. The force obtained was measured with the pinch gauge in kilograms and the average of the three results was calculated.

Box and Block Test (BBT): This test was designed to assess manual coordination and dexterity. The patient is seated in front of a wooden box divided into two identical halves containing 150 2.5 cm wooden blocks of different colors positioned in the midline. The patient’s task is to transfer as many blocks as possible from one half of the box to the other within 60 s [29]. The test is performed first with the dominant hand and then with the non-dominant hand. It is a standardized tool for measuring gross motor function of the upper limbs and has been validated for sex and age in healthy subjects [29]. It also shows low ceiling and floor effects in people with MS. Other psychometric properties, such as test-retest reliability and minimal detectable change, have been investigated in other neurological conditions and show excellent reliability and a cut-off point of 6 blocks for minimal detectable change [29].

Nine Hole Peg Test (NHPT): This test has been used to assess upper limb function, specifically fine motor skills of the hand [30]. The subject is positioned in front of a pegboard and is asked to insert and remove nine pegs, one at a time, into the nine holes of the board as quickly as possible. The dominant hand is tested first, followed by the non-dominant hand. The non-tested hand can provide stability by holding the pegboard. The score is measured in seconds and represents the time taken to complete the task. The NHPT has excellent test-retest, inter-rater, and intra-rater reliability, as well as adequate internal consistency. Scores greater than 0.27 s per pin indicate severe hand dysfunction. The minimum detectable change is 7.46 s for the non-dominant hand and 4.38 s for the dominant hand. The standard deviation of the measurement is 2.69 s for the non-dominant hand and 1.58 s for the dominant hand [30, 31].

Abilhand: It is a self-administered questionnaire designed to assess an individual’s manual dexterity, which is defined as the ability to perform a series of tasks regardless of the strategies used to perform them [32]. The questionnaire consists of 23 items that are scored on a scale of 0 to 2, corresponding to different activities of daily living (ADLs), and the patient indicates the level of difficulty experienced in performing the activities [33]. The score can range from 0 to 46, with higher scores indicating better manual dexterity. It has excellent inter- and intra-rater reliability, internal consistency, and construct validity [33].

Fatigue Severity Scale (FSS): It is a self-administered scale that assesses the severity of fatigue experienced by the patient during specific daily activities. It consists of 9 items and the scores range from 1 to 7, with a minimum score of 9 and a maximum score of 63. A higher score indicates a greater impact of fatigue on the patient’s life [34]. The standard error of measurement is 0.7 points, and the minimum detectable change is 1.9 points. Test-retest reliability is moderate, and the construct validity is excellent in people with MS [35].

Multiple Sclerosis Impact Scale (MSIS-29): It is a self-administered scale that assesses the impact of the disease on the patient’s life in the two weeks prior to administration. It consists of 29 questions, 20 of which assess the physical aspects of multiple sclerosis (MS) and 9 of which assess the psychosocial aspects. There are 5 possible answers, scored from from 1 to 5, indicating lower to higher impact of the disease. Two total scores are generated, corresponding to the physical and psychological impact subscales. The scores range from 0 to 100, with a higher score indicating a greater degree of disability [36]. The physical domain has been shown to have high internal consistency [37] and good construct validity, while the cognitive domain has good internal consistency [38]. The scale has low ceiling and floor effects, excellent Cronbach’s alpha for both parts of the scale, and a strong correlation between the physical and psychological domains [38].

Likert Satisfaction Questionnaire for the experimental group: At the end of the study, participants in the experimental group also completed a satisfaction survey about the experimental treatment using a Likert-type questionnaire developed by the research team. The questionnaire consisted of 20 items related to user satisfaction with the system used, including attributes of the tool itself, ease of use, accessibility, session design, schedules, duration, as well as attributes of the center and physiotherapist involved in the intervention. Scores could range from 1 to 5 points. The maximum possible score was 100 points, with higher scores indicating greater satisfaction.

Treatment Adherence and Adverse Effects: finally, the percentage of patients in both groups adhering to the treatment (%) and the possible occurrence of adverse effects were recorded.

Grip strength was selected as the main outcome. The effect size of grip strength was estimated to be medium (f = 0.3). A correlation of 0.5 between repeated measures was assumed. Considering three measurements (pre-intervention, post-intervention and one month follow-up), the sphericity correction was set at 0.5. With a statistical power of 0.80 and an alpha level of 0.05, a total sample size of 20 patients was estimated. Taking into account an attrition rate of 25%, a total of 25 patients were considered necessary. Sample size was calculated using G*Power 3.1.7 software.

All data were entered into the SPSS v.28.01 statistical package. Descriptive analysis of qualitative data was carried out using means, percentages, and ranges. The Kolmogorov-Smirnov test was used to assess the normality of the study variables. For variables with a normal distribution, a parametric analysis was performed using repeated measures analysis of variance (ANOVA) with Bonferroni post hoc adjustments. The group parameter was set as the between-subjects factor, while the within-subjects factors included each of the measurements and the treated side. A one-factor ANOVA was used to compare satisfaction and treatment attendance. Statistical analysis was performed at the 95% confidence level, with p-values less than 0.05 considered statistically significant.