Rehabilitation With and Without Robot and Allied Digital Technologies in Stroke Patients

Rehabilitation With and Without Robot and Allied Digital Technologies (RADTs) in Stroke Patients: a Pragmatic Multicenter Randomised Controlled Trial on the Effectiveness, Acceptability, Usability, and Economic-organizational Sustainability of RADTs From Subacute to Chronic Phase

Rehabilitation after stroke often employs Robot and Allied Digital Technologies (RADTs), yet evidence on their effectiveness remains inconclusive due to study heterogeneity and limited sample sizes.

This multicentre randomized controlled pragmatic trial aims at comprehensively and accurately assessing the effectiveness of rehabilitation mediated by RADTs after a stroke, compared to traditional rehabilitation, also considering economic sustainability.

The aim of this study is to evaluate, in a pragmatic trial on a large sample of patients who have experienced a stroke within the six months prior to enrollment, the effectiveness of a multimodal treatment using robotics and advanced technologies compared to traditional multimodal treatment, in the recovery of activities of daily living.

Primary Objective: To demonstrate, in a population of subacute stroke patients, the non-inferiority of a rehabilitation treatment integrated with robotic and/or technological devices compared to traditional rehabilitative treatment in the recovery of activities of daily living.

Secondary Objectives:

• To demonstrate the superiority of rehabilitative treatment integrated with robotic and/or technological devices compared to traditional rehabilitation treatment in the recovery of activities of daily living, should non-inferiority be demonstrated;
• To compare the improvements between two groups in all targeted domains (upper limb sensorimotor abilities; lower limb sensorimotor abilities and gait; balance; cognitive abilities), in accordance with the International Classification of Functioning, Disability, and Health (ICF);
• To analyze the neurophysiological parameters and factors involved in neuroplasticity processes;
• To compare the time pattern of manual dexterity and walking performance recovery in the two groups;
• To assess the effects of the rehabilitation treatment in terms of daily life activities and quality of life through medium-term follow-up;
• To evaluate the acceptability and usability of the rehabilitative treatment integrated with robotic and/or technological devices for the patient, their family, and the healthcare practitioner;
• To create a predictive model capable of forecasting the effectiveness of robotic and technological treatment in subjects after a stroke, based on a detailed description of the patient at baseline (using demographic data and clinical picture at admission);
• To assess the economic sustainability of the rehabilitative treatment integrated with robotic and/or technological devices for the patient, payer, and society, through the creation of a model for the assessment and prediction of cost-effectiveness, cost-utility, and, for the healthcare system, a Budget Impact Analysis, and related sensitivity analyses.

The study employs a multicenter, multimodal, randomized, controlled, parallel-group (1:1) interventional design with blinded assessors, following a pragmatic approach. It will be conducted across multiple clinical centers involved in a national research initiative, encompassing both outpatient and inpatient settings. Randomization will be centralized and stratified by clinical center, latency, and clinical setting (inpatients or outpatients).

The sample size (596 patients) was determined considering the following, with respect to the primary outcome (the change in the modified Barthel Index): (a) the non-inferiority of robotic treatment compared to traditional treatment; (b) a power of 80%; (c) a non-inferiority margin of 5 points; (d) a bilateral 95% confidence interval; (e) a standard deviation of the primary outcome of 20 points. Considering these parameters, a sample size of 506 patients (253 per group) is obtained. Additionally, considering a dropout rate of 15%, a total sample size of 596 patients is obtained. The same sample size is sufficient to also demonstrate the potential superiority of robotic treatment. In fact, considering a significance level of 5%, a power of 80%, a minimal clinically important difference of the scale of 9.25 points, a standard deviation of 20 points, and a dropout rate of 15%, a total of 178 patients are required for demonstrating superiority.

The experimental group will receive rehabilitation using RADTs within a new organizational model, where two physical therapists supervise four to six patients. In the control group, patients will undergo individual traditional rehabilitation, maintaining a 1:1 patient-to-therapist ratio. In both groups, patients will undergo comprehensive rehabilitation treatment, targeting the following domains: a) upper limb sensorimotor abilities; b) lower limb sensorimotor abilities and gait; c) balance; d) cognitive abilities.

The primary outcome is to demonstrate non-inferiority in activities of daily living as measured by the modified Barthel Index. Should non-inferiority be established, the study will then evaluate the potential superiority of RADTs in activities of daily living.

All participating centers will follow a standardized operating procedure regarding treatment and outcome assessment to ensure consistency across all sites. Data will be systematically collected using the REDCap (Research Electronic Data Capture) platform.

Inclusion Criteria:

• First-ever diagnosis of ischemic or haemorrhagic stroke confirmed by Computed Tomography or Magnetic Resonance Imaging;
• Age 18 years and over;
• Time since the event equal to or less than 6 months;
• Mild to severe impairment of the upper limb (motor section of the Fugl-Meyer Assessment of Upper Extremity ≤ 58) and/or mild to severe impairment of the lower limb (score on the Functional Ambulation Categories scale ≤ 4) and/or mild to severe impairment of balance (Berg Balance Scale ≤ 45);
• Clinical stability allowing transfer to the gym and execution of the planned treatments.

Exclusion Criteria:

• Clinical instability;
• Behavioral/cognitive disorders preventing adequate patient compliance with both traditional and robotic rehabilitation treatment (severe cognitive deficit, Montreal Cognitive Assessment <10);
• Rigidity or hypertonia (Modified Ashworth Scale > 3) in the plegic/paretic limb;
• Serious uncorrectable visual impairments preventing the patient from performing treatment with technological and/or robotic devices;
• Pregnant women;
• Refusal to sign the informed consent.