ARIS Armrobotics in stroke rehabilitation

Project managers

Linda Sørensen, Specialist Occupational Therapist – somatic health, MSc, Ph.d.-student Head of technological intervention center, Innovation, Sunnaas Rehabilitation Hospital HF. o 

Matthijs Wouda, Physiotherapist, Ph.d., Head of Clinical Physiological laboratory , Research Department, Sunnaas Hospital  HF, Associate professor  OsloMET.

Abstract

Stroke is one of the main causes of acquired disability in the adult population. The literature shows that 55% to 75% of stroke sufferers experience persistently reduced motor function in an arm, which is perceived as very disabling. The use of robotics in rehabilitation after an injury or illness, both as an aid and a training tool, will increase at record speed in the years ahead. The Norwegian health service must deal with constantly new challenges such as a shortage of health personnel and an introduction of technology in several arenas. Several rehabilitation institutions and municipalities want to use robotics in the training of their patients but lack experience and knowledge.  

The guideline has been developed through three phases: 1) Preparation of a systematic review, 2) Testing of a training and test protocol, 3) Interviews with patients and therapists.  

The results of the systematic review (18 randomised trials and 1,295 participants) showed that exercise with arm robotics improved arm and hand function to a greater extent than traditional therapy, both for those in the subacute and chronic phases after a stroke. The systematic review shows a significant positive effect of arm robotics on muscle strength and spasticity in the arm and hand, but not on performing daily activities. Due to great variation in dosage, number of participants and type of arm training robot in the included studies in the systematic review article, exercise dosing (i.e. how often and for how long arm robotics should be used) has been developed on the basis of literature and clinical experience/expertise.

Based on the systematic review and collaboration with the expert group, a training and testing protocol was developed with recommendations on exercise dosage, as well as which outcome measures (arm and hand function tests) should be used to measure the effect of an arm robotics intervention in people who have suffered a stroke. This protocol for training and testing with arm robotics was then tested on 19 patients in the subacute phase admitted to Sunnaas Hospital for rehabilitation after a stroke in the period September 2022 – May 2023.  

A total of 19 participants completed 4 weeks of robotics training with the new training protocol and they trained an average of 4 (min-max: 2.25-4.75) times a week. 17 out of 19 participants exercised at least 3 times a week. An average of 87.3% (min-max: 61-100%) of the scheduled sessions were completed. The number of scheduled training sessions over 4 weeks was individually adjusted. On average, participants were scheduled for a total of 18 (min-max: 13-20) training sessions over 4 weeks.  The participants were scheduled for training sessions of 45 minutes duration. Some workouts were shorter than planned. The average effective training time was 31 minutes (min-max 19-40 minutes per workout).

The interviews with the participants showed that they were overwhelmingly positive to the use of arm robotics in rehabilitation and stated that this was a fun way to train. Most participants found that 45 minutes was the appropriate length of training, but satisfaction varied somewhat depending on which robot was used. Robots with more gameplay options and more degrees of freedom was described as more motivating in terms of overall training session length, compared to robots with fewer degrees of freedom and less game selection. Most of the participants found that training up to 5 times a week worked well, some wanted more frequent training. The participants gave feedback that a good introduction to the various games/programs at the start was necessary to experience mastery, and to make good use of the time set aside for training. The majority of the respondents found that they received sufficient assistance during the organization with 1 therapist for 3 patients. Good competence in machines and adjustment possibilities was considered by both personnel and participants to be of great importance. Several were positive to the various adjustment possibilities that made it possible to exercise despite individually limited physical function at the start. The possibility of increasing the difficulty of the games as the feature changed was also highlighted as positive.

The interviews with the therapists showed that clinicians who offer arm robotics should have good knowledge of the machines' settings and adaptation possibilities in order for the patients to benefit ideally from the training. This competence is particularly important for adapting to the level of difficulty and adjustments along the way, as well as replacement/change of games to increase training outcomes, but also to maintain motivation in the individual. Several of the therapists also emphasized the importance of individual adaptation. Many reported that the training length of 45 minutes worked well, but that the choice of time should be adapted to function. A significantly reduced function can provide limited opportunities for variation and change in game/program, and that there may be a lot of repetitive work at some stages with high weight relief from the robot itself. The therapists state that grip strength measured by dynamometer, as well as the Box and Block test, are considered good mapping tools for evaluating changes in arm and hand function. The tests are also described as not very time-consuming tests that provide a lot of information. The Action Research Arm Test (ARAT) is considered by many to be appropriate for those who function across several joints.

Robot-assisted training can be offered as 'one-to-one training' or group training where a therapist/therapist assists three to four patients. The ARIS project has resulted in the following recommendations:

• People who have suffered a stroke in both the subacute and chronic phases may benefit from training with arm robotics.
• Arm robots with 1-7 degrees of freedom can be used in the rehabilitation of arm and hand function in patients who have suffered a stroke.  
• People who have undergone a stroke are advised to train 45 minutes, 3-5 times per week with arm robotics. The recommendation is indicative and individual adjustments (amount of training, type of arm robot, type of game, etc.) should be considered at all times.
• Personnel should have good knowledge and competence regarding equipment and games, as well as various possibilities for adjustment on each unit.  
• The training should be varied within the training sessions and across the training sessions in order to train on the right level of difficulty and maintain training motivation

 
Ethical approvals:

REK 278134

NSD 411964

Funding

DAM

Time period

Mars 2021 - December 2023