Integrated, Practice-oriented Electromechanical-assisted Gait Training in Subacute Stroke Patients

Background and Rationale:

Every year, more than a 15 million people worldwide suffer a stroke1. Despite laborious rehabilitation programs 70% of those stroke patients still show limited gait ability after three months.

Numerous RCTs have shown the superiority of over ground gait training and electromechanical-assisted gait devices in comparison with conventional over ground gait training alone. (Jan Mehrholz, Elsner, Werner, Kugler, \& Pohl, 2013) However, most of these trials used the electromechanical-assisted gait devices in a very high frequency which is almost impossible to achieve in daily clinical routine. Therefore, the effect of a practice-oriented physical therapy protocol including electromechanical-assisted gait training remains unclear.

Objective(s):

The aim of this clinical trial is to investigate the effects of an integrated, practice-orientated individual physical therapy protocol with or without electromechanical-assisted gait training in subacute stroke patients, respectively.

The effects of both physical therapy protocols will be compared with validated assessments covering the different components of the ICF framework: function, activity and participation.

Primary objectives: Effects on gait ability and on basic activities of daily living (ADL).

Secondary objectives: Effects on walking speed, walking capacity, functional mobility, muscle force, spasticity and reintegration to normal living (RNLI).

Furthermore, different training-related data are recorded to capture the efficiency (intensity and hassle) of both physical therapy protocols.

Detailed Description:

Background and Rationale:

With 16.9 million people suffered a first stroke and almost six million stroke-related deaths worldwide in 2010, stroke was ranked as the second most common cause of death at the age of over 65 years (Lozano et al., 2012). Although mortality-to-incidence ratios have decreased, the global burden of stroke in terms of the absolute number of people affected every year and stroke-related deaths are increasing (Krishnamurthi et al., 2013). Taking into account the demographic changes of our society up to 12 million stroke deaths and 70 million stroke survivors are expected in 2030 (Krishnamurthi et al., 2013).

In Switzerland 15'000 people suffer a stroke every year (Jorgensen, Nakayama, Raaschou, \& Olsen, 1995). Three months after stroke 20% of the affected patients are still bound to a wheelchair and 70% show limited gait ability, therefore, the restoration of gait ability is one of the most important therapy goals. Since motor restoration especially takes place during the first six months the choice of a proper gait therapy is of great importance (Jorgensen et al., 1995).

Since repetition and task-specific activity have been shown to be key factors for motor (re)-learning (Kwakkel, Kollen, \& Lindeman, 2004) (French et al., 2007) treadmill training with body weight support has been implemented in stroke rehabilitation. However, no superiority of treadmill training compared to conventional over ground gait training could have been shown to date. Especially patients who cannot walk at all don't seem to benefit from treadmill training to a greater extent (J Mehrholz et al., 2014). A possible explanation therefor is the tremendous therapeutic effort needed to move severely impaired patients on a treadmill. To obtain a higher therapeutic efficiency (reduce therapeutic effort and increased number of steps) electromechanical-assisted gait devices have been developed. During electromechanical-assisted gait therapy the patient is either moved by the means of a robotic-driven exoskeleton orthosis (exoskeleton device) (Colombo, Joerg, Schreier, \& Dietz, 2000) or two mechanically driven foot plates simulating the gait pattern (end-effector device) (Hesse, Sarkodie-Gyan, \& Uhlenbrock, 1999).

Over the last decade different studies have evaluated the effect of the implementation of the exoskeleton and end-effector devices within the rehabilitation process of stroke patients (Jan Mehrholz et al., 2013). RCTs have shown the superiority of the combination of conventional over ground gait training supplemented by electromechanical-assisted gait devices in comparison with conventional over ground gait training alone (Jan Mehrholz et al., 2013). Especially non-ambulatory patients in the subacute phase (< 3 months after stroke) benefit from this type of intervention. Furthermore, the choice of the right gait device matters. A sub analysis showed a statistically significant difference in the recovery of walking speed in favor of the end-effector compared to exoskeleton devices (Jan Mehrholz et al., 2013). The largest study ever performed with an end-effector device, "Deutsche Gangtrainer Studie (DEGAS)" showed that a training program of 20 therapy sessions each consisting of 20 minutes walking on an end-effector device combined with 25 minutes of conventional physiotherapy more than doubles the chance of regaining an independent gait ability compared to the same dosage of conventional physiotherapy alone (Pohl et al., 2007). Furthermore, the use of electromechanical-assisted gait therapy is safe and does not result in more adverse events (AE) than conventional physiotherapy (Jan Mehrholz et al., 2013).

Many of the performed RCTs with electromechanical-assisted gait therapy have chosen a high training intensity of up to five training sessions per week. According to those findings the German Society for Neurorehabilitation (DGNR) recommends in the so-called S2e-Guidelines "Rehabilitation der Mobilität nach Schlaganfall (ReMoS)" a training dosage of 500-1000 steps per day for non-ambulatory stroke patients (Loipl \& Wittenberg, 2015). In daily clinical routine however, it is almost impossible to achieve such a high training intensity due to limited resources. Thus, it remains unclear whether a realistic therapy program including around three electromechanical-assisted gait training sessions per week leads to the same beneficial effects on gait ability. Therefore, the aim of this clinical trial is to investigate the effects of electromechanical-assisted gait training within an integrated, practice-orientated physical therapy protocol, which is compared to the effects of the same amount of physical therapy without electromechanical-assisted gait training in subacute stroke patients.

Study 0verall Objective:

The aim of this clinical trial is to investigate the effects of an integrated, practice-orientated individual physical therapy protocol with respectively without electromechanical-assisted gait training in subacute stroke patients.

Primary Objective:

To compare the effect of both physical therapy protocols with respect to gait ability and competence in basic ADL.

Secondary Objectives:

To compare the effect of both physical therapy protocols with respect to walking speed, walking capacity, functional mobility, muscle force, spasticity, and RNLI.

As further objectives, different training-related data are recorded to capture the efficiency (intensity and hassle) of both physical therapy protocols.

Safety Objectives The risks of this study are kept to a minimum. No specific safety objectives have been defined.

Primary Study Outcome:

The changes in gait ability and the competence in basic ADL are determined as primary outcomes. Both are assessed before and after the four-week intervention period (to proof short term effects) as well as three months after the end of the intevention period (3months FU).

Gait ability and competence in basic ADL are regarded as to be the most important therapy goals of subacute stroke patients. Furthermore, these outcomes were also used in other clinical trials with electromechanical-assisted gait training making results comparable.

Gait ability (ICF-level activity):

To assess gait ability the FAC is performed. The FAC is technically simple, inexpensive, reliable and sensitive to change in the assessment of patients after stroke who cannot walk without personal assistance at the beginning of their inpatient rehabilitation. The FAC provides information about the way of physical (not technical e.g. walking aids, orthosis) assistance that is needed for walking.

Competence in basic ADL (ICF-level activity):

To assess competence in basic ADL the FIM is performed (ANQ, 2013). The FIM shows an excellent test-retest-reliability (Hobart et al., 2001), an excellent internal consistency, excellent criterion validity and high responsiveness (for the motor subscale) in stroke patients (Jan Mehrholz, Wagner, Rutte, Meißner, \& Pohl, 2007). The nurses from ZRW assess the FIM in every rehabilitation service unit on a weekly basis(Tyson \& Connell, 2009). Therefore, a high level of standardisation is given.

Secondary Outcomes:

The changes in walking speed, walking capacity, muscle force, spasticity, functional mobility and RNLI are determined as secondary outcomes. All secondary outcomes except RNLI are assessed before and after the four-week intervention period. RNLI is measured at the 3months FU.

Walking speed (ICF-level activity):

To assess walking speed the 10mWT is performed. The 10mWT is a simple, highly reliable and valid test for the assessment of walking speed in stroke patients (Tyson \& Connell, 2009). It is carried out in patients with a FAC-level >=2. Lower FAC-levels are not tested.

Walking capacity (ICF-level activity):

To assess walking capacity the 6minWT is performed. The 6minWT is a simple and valid test which is used to assess maximum walking distance. It shows adequate reliability and excellent validity in stroke patients (Kosak \& Smith, 2005). The 6minWT is carried out in patients with FAC-Level >=2. Lower FAC-levels are not tested.

Functional Mobility (ICF-level activity):

For the assessment of mobility, the DEMMI is chosen. The DEMMI consists of 17 basic everyday mobility tasks differentiating between 5 conditions (bed, chair, static balance, walking and dynamic balance) (de Morton, Davidson, \& Keating, 2008). The DEMMI is a practical, short and valid mobility test for geriatric and stroke patients that is sensitive to detect changes (Braun et al., 2015).

Muscular Strength lower extremities (ICF-level structure/function):

Change of lower extremities muscular strength is measured with the Motricity Index (MI). Due to the present intervention studied only the muscle force of the lower limbs will be evaluated (hip flexion, knee extension, ankle dorsiflexion). Previous studies have shown that the MI is a valid, reliable and sensitive tool that is short and easily to apply (Collin \& Wade, 1990; Demeurisse, Demol, \& Robaye, 1980).

Spasticity (ICF-level structure/function):

Quantification of spasticity of the affected lower limb is carried out with the mTS. The mTS has shown to have very good test-retest reliability and sensitivity that is high enough to measure changes pre-post treatment (Jan Mehrholz et al., 2005).

Reintegration to normal living (ICF-level participation):

For the assessment of the ICF level participation the RNLI is chosen. The instrument can be used to evaluate the degree to which the patient has been able to return to a normal life (Wood-Dauphinee \& Williams, 1987).