Study design and patient population

The theRapeutic effects of peroneal nerve functionAl electrical stimuLation for Lower extremitY in patients with convalescent poststroke hemiplegia (RALLY) study is a randomised, controlled, open-label multicentre study. The trial enrols patients with hemiplegic gait disorder (drop foot) due to stroke from 21 poststroke rehabilitation centres across Japan. The major inclusion criteria are patients aged 20–85 years, who had an initial stroke ≤6 months prior to the enrolment, with functional ambulation classification (FAC) 3 or 4. The FAC categorise patients according to basic motor skills necessary for functional ambulation. The score ranges from 0 to 5; FAC 3 patients have a walking disability that requires stand-by assistance, whereas patients classified as FAC 4 can walk independently on level ground but requires support on stairs. Based on previous literature, these are the patients who are likely to benefit from training with FES. Those patients who cannot tolerate rehabilitation due to existing conditions such as heart failure or severe osteoarthritis are excluded. For safety reasons, very elderly patients (>85 years), patients with unstable blood pressure, heart rate or blood glucose levels, or patients with severe liver, kidney and cardiovascular dysfunction, are excluded from the study. Patients with other conditions that may render the outcome are also excluded (eg, use of FES or a robot suit within 1 month prior to providing consent or botulinum toxin injections or phenol nerve block injection within 6 month of informed consent, severe sensory dysfunction or higher brain dysfunction). Details of the inclusion and exclusion criteria are listed in box 1. Physician investigators at each participating site are responsible for recruiting patients and obtaining written informed consent. Subsequently, candidate patients were then tested for FES receptivity for a maximum of 7 days (screening period) by the treating physician and physical therapists. Patients were excluded from the study if any of the followings are observed: (1) unresponsive to the FES device, (2) intolerance to continuous stimulation or (3) if gait function improves significantly during the screening period.

After the screening period, eligible patients were randomly sorted to receive either FES training using Walkaide (FES group) or to rehabilitation without FES (control group), in addition to their usual training. Gait abilities and functions will be assessed at each participating site within 3 days after randomisation (pretreatment period) and at the end of the study. The details of the schedule and assessments are summarised in figures 1 and 2.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Study schema. Randomised patients undergo 8 weeks of allocated treatment in addition to the usual rehabilitation programme. *Patients who already use AFO at the time of randomisation will perform each test with AFO. AFO; ankle-foot orthosis; FES, functional electrical stimulation; 6-MWT, 6 min walk test.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Study flow. Enrolled patients undergo 1 week of screening period to confirm tolerability and receptivity of FES. FES, functional electrical stimulation.

Data entry is being performed by either the institutional principal investigator (physician) or the subinvestigators (physician or physical therapists) using an electronic data capturing system and is monitored centrally per protocol to promote data quality.

In case of research-related harm, the participating site will be responsible for providing adequate treatment and compensation, as required. Although we do not anticipate any severe side effects, this is a required rule for any trials performed in Japan.

Patient and public involvement

The research question was developed from patients’ experiences and needs. The patients were not involved in the design, recruitment or conduction of this study; however, they all consented to the research question and outcome measures. The burden of the intervention was assessed by the patients themselves at the run-in period, and the results of the study will be disseminated to the patients by their treating physicians.

Randomisation

The enrolled patients are randomly allocated to either the FES or control group in a 1:1 fashion with a minimised method using an electronic data capturing system, the eClinical Base (https://www.tri-kobe.org/support/tools/). Randomisation was performed with consideration to the following factors: FAC 3 or 4, age<65 years or not, types of stroke (ischaemic stroke or haemorrhagic stroke) and institution. The institutional principal investigators were responsible for creating an anonymous number control table and for securely concealing the sequence.

Treatment programs

The usual rehabilitation training is defined as a 60 min physiotherapy treatment provided 5 days a week to both the FES and control groups, over 8 weeks, for a total of 40 days. These are based on the Japanese insurance policy, and it is the clinically adopted routine in Japan. The programme consists of basic activity training, including mat exercise, standing up and sitting down and ambulation with assistive devices or support and/or range of motion (ROM) training and/or gait training using an AFO (if the patient is already using it at the time of recruitment).

In addition to the usual rehabilitation training, the patients included in this study receive their allocated programme (FES or control). According to Japanese insurance policy, a 40 min training programme is considered one unit; all the patients undergo one training unit per day over an 8-week interventional period (figure 1). Any rehabilitation programmes initiated before this trial are continued under the condition that intervals, duration or contents remain the same throughout the trial.

FES group (intervention group)

The FES device used in this study is Walkaide (Innovative Neurotronics). The researchers set up the surface electrode onto the proximal region of the lower limb with the cuff. An internal sensor detects leg tilt and determines whether an individual is in the swing or stance phase, then stimulates the common peroneal nerve in the swing phase. The stimulation from FES activates the weakened muscle to lift the foot appropriately and, at the same time, assists with recovery by restoring the nerve-to-muscle signals.

The TES (therapeutic electrical stimulation) mode is used in patients with FAC category 3, while the HAND mode (manual electrical stimulation) and TILT mode (electrical stimulation delivered in the swing phase based on a tilt sensor) are used in patients with FAC category 4. Treatment modes are selected and adjusted by qualified programme providers. In the TES mode, electrical stimulation is repeated at regular intervals. This mode is used for muscle stretching and joint range training. The stimulation and pause times for the TES mode are programmed prior to rehabilitation. In the HAND mode, physicians or physical therapists manually control the FES using a hand switch. In TILT mode, FES operates automatically according to the incline of the foot. Walkaide senses the tilt and automatically adjust the mode so that it stimulates the common peroneal nerve to lift the foot at the right time during the gait cycle, prompting a more natural, efficient and safe walking pattern.

The use of AFO is prohibited during the FES training. All the training is overseen by the physician or physical therapists.

Control group

Training is provided to the control group without using the FES. Training for patients with FAC category 3 includes self-stretching and foot dorsiflexion ROM training (triple foot triceps stretch training to extend the foot dorsiflexion ROM); training for patients with FAC category ≥4 includes gait training using an AFO.

Outcomes

The primary outcome measured will be the change in the 6 min walk test (6-MWT) defined as the difference in distance during a 6 min timed walk (metres) barefoot at weeks 0 and week 8. 6-MWT is a feasible measure that is often used to determine functional capacity in individuals with compromised ability and has been shown to be highly reliable in evaluating various aspects of gait performance in individuals with chronic mild to moderate hemiparesis after stroke.21–23 Secondary outcomes included changes in barefoot gait speed as assessed using the 10 m walk test.6 24 Walking speed and endurance are two important aspects of walking capacity in patients with stroke. The 10 m walk test is performed at a comfortable walking speed, and the average value of two measurements is calculated.10 11 25 Also, as secondary outcomes, the 6-MWT and 10 m walk test will be evaluated using WA with AFO, if relevant. In addition, the difference in the total Fugl-Meyer assessment score (only the lower extremity part relevant to this trial) will be evaluated to assess the physical recovery poststroke. The modified Ashworth scale scores are used to assess spasticity. The Ashworth scale is a widely used clinical measure of joint resistance.26 The Ashworth scale subjectively grades the manual sensation of mechanical resistance experienced by the examiner during a one second joint rotation over the full ROM. Furthermore, the differences in active and passive ROM for ankle dorsiflexion, timed up-and-go test score (by barefoot, using FES and AFO-assisted, if relevant), the Stroke Impact Scale score, the patient-reported outcome measure and the videotaped gait assessment will be evaluated. Passive and active ankle-dorsiflex on ROM are measured by the physical therapist or physician and are recorded in 5° increments. Stroke Impact Scale covers eight domains: strength, hand function, mobility, activities of daily living, memory, communication, emotion and handicap and is an established instrument to assess stroke recovery.27 28 The patient-reported outcome measure has been widely used in rehabilitation trials in Japan to capture patient perception of their own health status or quality of life without interpretation by clinician. It consists of three questionnaires: (1) patient-reported burden in raising the foot during bare-foot walking, (2) patient-reported spasticity while walking bare-footed, (3) patient-reported stability in bare-footed walking. Ten-metre walk tests are videotaped and gait disturbance is evaluated by an independent central adjudication committee which comprised of doctors, physical therapists and occupational therapists using the Rivermead Visual Gait Assessment, which is a four-point scale visual gait assessment form validated elsewhere.29 For safety assessment, any adverse event will be collected regardless of its severity (box 1). All the outcomes are collected at week 0 (pretreatment period) and week 8, except for adverse events which will be collected at any the time of the event.

Study organisation

Periodic assessments of safety and efficacy are performed by an independent Data Monitoring Committee (DMC) composed of three members with appropriate expertise. Project and data management, and data analysis are conducted by the Translational Research Informatics Center for Medical Innovation (http://www.tri-kobe.org/). The monitoring is done centrally to ensure that the randomised cases reach their target number and that the study is executed per protocol. In the case of serious protocol deviation, serious adverse events or safety issues, a site visit may be performed whenever necessary. Audits will be performed to abide by Japanese laws and will follow a discrete Standard Operating Procedure independent from the investigators and the sponsor.

A protocol amendment will be required if there is any change in study design, eligibility criteria, intervention, endpoint, endpoint analyses or targeted event during the study. The amendment will need to be approved by the institutional review board, and the study principal investigator is responsible for communicating changes to institutional investigators, the data monitoring committee, the data centre and the study statisticians.

Sample size estimates

Sample size was calculated based on feasibility. With 200 cases equally randomised to the control and intervention group, a difference of 43.8 m in the 6-MWT is detectable with statistical power of 80%, on the assumption that the SD is 110 m.

Statistical analyses

Patient characteristics will be summarised descriptively for quantitative variables and as a number and percentage for qualitative variables. The change in 6-MWT distance before and after the treatment will be evaluated for all the participants. Analysis of covariance will be performed to compare the change in the 6-MWT distance between the groups and to estimate the adjusted difference and 95% CIs. Covariates included in the analysis will be FAC, age and type of stroke. Since this is a prospective study with central data monitoring system, targeting in-hospital patients, a high proportion of non-adherence cases and/or missing data are unexpected.

Access to the final trial data set and trial results

The authors will have access to the final trial data set; the trial results will be presented to the public at a professional conference and in a peer reviewed journal.