Abstract
Background and Objectives:
Robot-assisted gait training (RAGT) constitutes a modern neurorehabilitation approach. However, limited data are available regarding the efficacy of RAGT when combined with conventional rehabilitation in the early subacute phase of stroke. This pilot study assessed the additive impact of protocol-defined RAGT on functional ambulation categories (FACs), walking abilities, and balance.
Methods:
This randomized controlled trial used a 1:1 Prospective Randomized Open Blinded End-point design (NCT04910217), comparing two groups: conventional rehabilitation and the RAGT group. All participants received protocol-defined conventional rehabilitation (15 sessions over 3 weeks). The RAGT group underwent additional sessions using the Lokomat Pro FreeD (15 sessions over 3 weeks). FAC, 10-meter Walk Test (10MWT), Timed Up and Go (TUG), and Berg Balance Scale (BBS) were assessed at 3 weeks and 3 months after randomization. In addition, the modified Rankin Scale was evaluated.
Results:
Among 184 patients screened between June 1, 2020, and November 30, 2023, 42 were included (19 in the RAGT group and 23 in the conventional group). The median time from stroke to randomization was 13 days (interquartile range [IQR]: 11–16), and the median age was 67 years (IQR: 58–71). Both groups demonstrated improvements in FACs. Significantly, patients in the RAGT group showed better results in the endpoint of FAC improvement by 2 or more points (P = 0.044). Progress in TUG (P = 0.034) and BBS (P = 0.049) during the study period was significantly more favorable in the RAGT group.
Conclusions:
The addition of RAGT to conventional rehabilitation in the subacute phase of stroke exhibited positive effects on improving gait, walking, and balance.
Trial registration number: NCT04910217
Keywords: Stroke, robot-assisted gait training, rehabilitation
Introduction
Recent research indicates that over 20% of individuals with stroke fail to recover their walking function, and 70% exhibit a diminished walking speed.[1,2] Merely 30%–50% of post-stroke patients achieve community ambulation, a key metric in the activities and participation domains of the International Classification of Functioning, Disability, and Health.[3] In line with core neurorehabilitation principles, enhancing neuroplasticity through motivation, early initiation, multisensory stimulation, and appropriately timed, progressively intensified therapy is essential for effective gait training, especially when task-specific.[4] Robotic devices and exoskeletons offer promising means of delivering earlier and more targeted rehabilitation than conventional methods.[5] While conventional rehabilitation approaches can be effective, human resource limitations may limit the delivery of intensive, high-repetition training protocols that could optimize neuroplastic changes in stroke patients. For instance, therapists may be unable to effectively control the offload weight and segmental weight, which would facilitate task-oriented training. A body of moderate-quality evidence suggests that including gait-training devices alongside conventional rehabilitation could be advantageous for post-stroke patients, facilitating their capacity to initiate independent walking.[6,7,8,9]
Conventional rehabilitation employs diverse concepts and techniques focused on enhancing specific tasks, including postural trunk stability, balance, and gait stability.[2,10] The first ten post-stroke weeks have been identified as crucial for gait recovery.[2] Robot-assisted gait training (RAGT), utilizing exoskeletons for post-stroke patients, embodies a contemporary neurorehabilitation approach for the early stages of rehabilitation. Nevertheless, clear and high-quality recommendations regarding the optimal regimen for RAGT training, including intensity and timing, are currently lacking. Consequently, these aspects remain subjects of ongoing clinical research.[7,9,11] Clinical trials assessing the combined impact of RAGT with conventional rehabilitation indicate a higher likelihood of patients regaining independent walking within a 6-month post-stroke period in the RAGT group. Notably, a recent Cochrane Review noted that the advantageous effects of RAGT were more pronounced in patients who were initially nonambulatory.[9] However, randomized studies focusing on the outcomes of early RAGT of hospitalized are currently lacking, which poses a gap in the current literature.
Gait and its associated parameters can be assessed through standardized scales, with functional ambulation category (FAC), the 10-meter Walk Test (10MWT), the Timed Up and Go (TUG) test, and the endurance characteristics of walking (specifically, the 2- and 6-min walking tests) being commonly employed for such evaluations.[1,9]
In the present pilot randomized study, we aimed to investigate the impact of the addition of 3-week RAGT in the very early phase after ischemic stroke on the (i) functional status measured by the FAC and modified Rankin scale (mRS), (ii) temporospatial parameters of gait (10MWT and TUG tests), and balance (Berg Balance Scale [BBS]).
Methods
This randomized single-center trial was conducted on hospitalized patients at the Department of Rehabilitation and Physical Medicine at the University Hospital Ostrava in the Czech Republic, between June 2020 and November 2023. The study was registered (2022_05_01) at ClinicalTrials.gov (NCT04910217), https://clinicaltrials.gov/, and the study protocol was published previously.[12] The study was approved by the Institutional Ethics Committee of the University Hospital Ostrava (reference number 395/2020). Written informed consent to participate in the study was obtained prior to randomization.
Participants and setting
This study involved patients diagnosed with their first-ever ischemic stroke confirmed by brain magnetic resonance imaging who were hospitalized at the Department of Rehabilitation and Physical Medicine (admitted directly from the Stroke Unit) at the University Hospital Ostrava in the Czech Republic.
Inclusion criteria were: age ≥18 years, time from stroke to first rehabilitation <6 weeks (aiming for inclusion as soon after the stroke as possible), signed informed consent, early mRS 2–4 (pre-stroke mRS 0–1), early FAC 0–3 (pre-stroke FAC 5), ability to stand with support for 3 min, and vertical tolerance >15 min.
Exclusion criteria included prior ischemic or hemorrhagic stroke or other known/new brain disease (e.g., tumor, multiple sclerosis, brain/spinal injury); serious comorbidities impairing long-term rehabilitation or causing chronic/progressive gait disorders; limited cooperation (e.g., MoCA <20, fear of LOKOMAT, sensory/cognitive deficits); impaired skin integrity on the lower torso/limbs precluding exoskeleton use; and limitations related to the device (weight >135 kg, thigh length outside 23–35 cm, or shank length outside 35–47 cm).
A target group size of 40 was based on prior RAGT studies showing significant differences in FAC outcomes between groups.[13,14,15] This pilot study did not include a priori sample size calculation due to the lack of available data on expected effect sizes for FAC improvements ≥2 points in the very early subacute phase of stroke rehabilitation. The aim of this study was to explore the potential clinical effect of adding RAGT to early conventional rehabilitation.
Randomization
Following screening and baseline assessments, eligible participants were randomly assigned (1:1) to the RAGT or conventional groups using stratified block randomization with a block size of 16. A pre-prepared table guided sequential patient allocation. The physician enrolling patients was blinded to group assignment. Stratification was based on baseline FAC: Group A included FAC 0–1, Group B FAC 2–3.
Interventions
Conventional rehabilitation involved a multidisciplinary approach including physiotherapy, occupational therapy, and mechanotherapy (MotoMed Viva 2 Letto), tailored to individual needs. Therapy was delivered by personnel trained in neurophysiological methods such as Bobath, proprioceptive neuromuscular facilitation, and dynamic neuromuscular stabilization. Sessions occurred 5 days a week: occupational therapy 30–45 min, physiotherapy 30–60 min, and mechanotherapy 20–30 min daily, totaling 15 sessions during hospitalization.
The RAGT group received additional Lokomat Pro FreeD® robot-assisted gait therapy five times weekly, 20–50 min per session, totaling 15 sessions. Supervised by trained physiotherapists, sessions ensured safe, high-quality gait, with parameters adjusted to optimize physiological walking. The program followed a standardized three-step process—safe walking, physiological gait, and goal-oriented gait—using progressive training principles to increase time, speed, steps, body-weight support, and guiding force. Lokomat Pro FreeD supported speeds of 0.5–2.5 km/h, with all training data recorded automatically for analysis.
Outcome assessment
Baseline scores were evaluated by the physiotherapist working with the patient. Follow-up assessments were performed by a different therapist (Prospective Randomized Open Blinded End-point). All outcomes were assessed before randomization (baseline), after 15 days of training, and 3 months post-randomization; the only exception was mRS, which was scored before randomization, at 3 weeks, and at 3 months from the index event.
FAC, ranging from 0 (unable to walk) to 5 (independent walking on all surfaces) at 3 months, was used as the primary outcome (see Supplementary Table 1 for more details on FAC). The secondary outcomes included, in addition to FAC after the 15th day of therapy, also mRS, 10MWT, TUG, and BBS (all after the 15th day of therapy and after 3 months). Changes in these parameters were evaluated using paired tests as appropriate.
Supplementary Table 1.
Functional Ambulation Category – baseline, after 3 weeks and 3 months, evaluation of the improvement
| RAGT |
Conventional |
p | |||||
|---|---|---|---|---|---|---|---|
| Baseline (n=19) | 3 weeks (n=17) | 3 months (n=17) | Baseline (n=23) | 3 weeks (n=20) | 3 months (n=19) | ||
| FAC | |||||||
| 0 | 4 (21.1) | --- | --- | 4 (17.4) | --- | --- | |
| 1 | 6 (31.6) | 1 (5.9) | --- | 7 (30.4) | 4 (20.0) | 1 (5.3) | |
| 2 | 5 (26.2) | 4 (23.5) | 1 (5.9) | 5 (21.8) | 1 (5.0) | 3 (15.8) | |
| 3 | 4 (21.1) | 4 (23.5) | 1 (5.9) | 7 (30.4) | 9 (45.0) | 2 (10.5) | |
| 4 | --- | 6 (35.3) | 9 (52.9) | --- | 5 (25.5) | 8 (42.1) | |
| 5 | --- | 2 (11.8) | 6 (35.3) | --- | 1 (5.0) | 5 (26.3) | |
| Improvement | |||||||
| Baseline to 3 months | 17/17 (100.0%) | 19/19 (100.0%) | --- | ||||
| Baseline to 3 weeks | 17/17 (100.0%) | 15/19 (78.9%) | 0.106 | ||||
| 3 weeks to 3 months | 10/17 (58.8%) | 13/19 (68.4%) | 0.730 | ||||
RAGT: robot-assisted gait training, FAC: functional ambulantion category
The mRS assesses functional independence and represents a standard outcome measure in post-stroke patients (range from 0, i.e., complete post-stroke recovery, to 6, i.e., death).[16] The mRS was assessed by a neurologist specializing in stroke care, who was blinded to the allocation. For mRS, in addition to this standard evaluation, a dichotomized evaluation into good outcome (1-2) and poor outcome (3-4) was performed.
The 10MWT (m/s) measures walking speed over 10 m from standing, while the TUG (seconds) records time to stand, walk 3.5 m, turn, and sit.[17] Three repetitions were done per time point, with the median used for analysis. If patients required assistance, gait velocity was recorded as zero and parameters were not assessed.
The BBS includes 14 items assessing balance during standing, sitting, transfers, and rotations—key daily activities. Each item is rated from 0 (no performance) to 4 (best), with a maximum score of 56 indicating stability. Interval scores can be categorized accordingly. Initially developed to predict falls in the elderly, the BBS is also useful for assessing balance in stroke patients.[17]
Statistical analysis
Numerical variables are expressed as medians and interquartile ranges, categorical variables as absolute and relative frequencies (%). Primary endpoints were FAC improvements by at least two points over 3 weeks and 3 months, with secondary analyses of FAC improvements by 1, 3, and 4 points. Secondary outcomes included mRS improvement (absolute and dichotomized) and changes in 10MWT, TUG, and BBS between time points. Between-group differences were analyzed using Mann–Whitney or Fisher’s exact tests as appropriate. Paired Wilcoxon tests assessed progress from baseline to 3 weeks and 3 months. Analysis was conducted using R software (v4.2.3) with maximal available data per time point. Significance was set at 0.05.
Results
A total of 184 ischemic stroke patients were screened between June 1, 2020, and November 30, 2023. Of these, 141 were excluded (99 did not meet criteria, 42 declined). Forty-two patients (33% women) met criteria and were randomized: 19 to the RAGT group and 23 to the conventional group. Follow-up was completed by 17 RAGT and 19 conventional patients, as shown in the Consolidated Standards of Reporting Trials (CONSORT) diagram [Figure 1], illustrating recruitment and dropouts. Two patients in each group tested positive for Coronavirus Disease 2019 (COVID-19), leading to premature rehabilitation termination; one RAGT patient had a recurrent stroke, and another died following COVID-19 infection. Baseline demographics and stroke treatment data are presented in Table 1.
Figure 1.
The study flow chart. (CONSORT: Consolidated Standards of Reporting Trials). RAGT: robot-assisted gait training
Table 1.
Demographic and treatment data
| Total n=42 | RAGT n=19 | Conventional n=23 | p | |
|---|---|---|---|---|
| Age, years, median (IQR) | 67 (58-71) | 63 (55-68) | 69 (64-71) | 0.126 |
| Sex, female, n (%) | 14 (33.3) | 5 (26.3) | 9 (39.1) | 0.515 |
| Metabolic syndrome, n (%) | 23 (54.8) | 9 (47.4) | 14 (60.9) | 0.535 |
| History of hypertension, n (%) | 35 (83.3) | 13 (68.4) | 22 (95.7) | 0.034 |
| Diabetes mellitus, n (%) | 20 (47.6) | 5 (26.3) | 15 (65.2) | 0.016 |
| Localization of ischemic stroke, n (%) | >0.999 | |||
| Anterior circulation | 26 (61.9) | 12 (63.2) | 14 (60.9) | |
| Posterior (vertebrobasilar) circulation | 16 (38.1) | 7 (36.8) | 9 (39.1) | |
| Stroke treatment, n (%) | 0.756 | |||
| Recanalization treatment (IV tPA±mechanical thrombectomy) | 18 (42.9) | 9 (47.4) | 9 (39.1) | |
| Conservative | 24 (57.1) | 10 (52.6) | 14 (60.9) | |
| FAC–Baseline, n (%) | 0.925 | |||
| 0 | 8 (19.0) | 4 (21.1) | 4 (17.4) | |
| 1 | 13 (31.0) | 6 (31.6) | 7 (30.4) | |
| 2 | 10 (23.8) | 5 (26.2) | 5 (21.8) | |
| 3 | 11 (26.2) | 4 (21.1) | 7 (30.4) | |
| Modified Rankin Scale – Baseline, n (%) | 0.750 | |||
| 3 | 15 (35.7) | 6 (31.6) | 9 (39.1) | |
| 4 | 27 (64.3) | 13 (68.4) | 14 (60.9) | |
| Days since onset to randomization, median (IQR) | 13 (11-16) | 13 (12-17) | 12 (12-16) | 0.799 |
FAC: functional ambulation category, IQR: interquartile range, RAGT: robot-assisted gait training, IV tPA: intravenous tissue plasminogen activator. The values represent absolute and relative frequencies (%). The p value was obtained using the Mann–Whitney test or Fisher’s exact test
Primary outcome
The primary outcome (3-month FAC improvement by ≥2 points) was achieved by 16 of 17 (94.1%) patients in the RAGT group versus 12 of 19 (63.2%) in the conventional group [Figure 2], a significant difference (P = 0.044, Fisher’s exact test). Other FAC analyses [Figure 2] showed similar trends but no significant differences (note: only 27 patients had complete data at all time points to assess improvement by ≥3 FAC points). All patients improved by at least 1 FAC point over 3 months. This improvement occurred within the first 3 weeks in all RAGT patients, compared to 78.9% in the conventional group. Supplementary Table 1 showed detailed FAC progression (baseline, 3 weeks, 3 months) in both groups. Improvement between weeks 3 and 12 appeared more marked in the RAGT group and persisted after RAGT cessation. Specifically, ≥2-point FAC gains were seen in 6 of 9 RAGT patients vs. 3 of 14 in the conventional group during this interval, though not statistically significant.
Figure 2.
Visualization of FAC improvement from baseline to 3 weeks (left) and 3 months (right). The RAGT group is dark grey, the conventional group light grey. Columns reflect maximum possible improvement per patient, considering the FAC ceiling score of 5 (e.g., a baseline FAC of 3 limits improvement to 2 points). FAC: functional ambulation category, RAGT: robot-assisted gait training
Secondary outcomes
Table 2 presents detailed mRS outcomes. In the RAGT group, 16 out of 17 patients (94.1%) followed up until the end of the study improved at 3 months. Of this number, 12 (75.0%) improved by ≥2 mRS points. In the conventional group, only 14 out of 19 patients (73.7%) improved, of which 9 (64.3%) improved by ≥2 points. However, in view of the relatively modest sizes of the groups, these differences were not statistically significant. The lower part of Table 2 dichotomizes mRS scores into 1–2 (mild disability) vs. 3–4 (moderate disability), providing less detailed but clinically meaningful differentiation. The results of this analysis mirrors the findings found in the analysis of the individual mRS categories—i.e., favorable trends for RAGT without statistically significant differences.
Table 2.
Modified Rankin scale–baseline, after 3 weeks, and after 3 months
| RAGT |
Conventional |
p | |||||
|---|---|---|---|---|---|---|---|
| Baseline n=19 | 3 weeks n=17 | 3 months n=17 | Baseline n=23 | 3 weeks n=20 | 3 months n=19 | ||
| mRS category, n (%) | |||||||
| 1 | --- | --- | 5 (29.4) | --- | 1 (5.0) | 5 (26.3) | |
| 2 | --- | 7 (41.2) | 8 (47.1) | --- | 4 (20.0) | 7 (36.7) | |
| 3 | 6 (31.6) | 6 (35.3) | 3 (17.6) | 9 (39.1) | 9 (45.0) | 3 (15.9) | |
| 4 | 13 (68.4) | 4 (23.5) | 1 (5.9) | 14 (60.9) | 6 (30.0) | 4 (21.1) | |
| Improvement | |||||||
| Baseline to 3 months | 16/17 (94.1%) | 14/19 (73.7%) | 0.232 | ||||
| Baseline to 3 weeks | 12/17 (70.6%) | 10/19 (52.6%) | 0.322 | ||||
| 3 weeks to 3 months | 11/17 (64.7%) | 11/19 (57.9%) | 0.742 | ||||
| Dichotomized mRS, n (%) | |||||||
| 1–2 | --- | 7 (41.2) | 13 (76.5) | --- | 5 (25.0) | 12 (63.0) | |
| 3–4 | 19 (100.0) | 10 (58.8) | 4 (23.5) | 23 (100.0) | 15 (75.0) | 7 (37.0) | |
| Improvement | |||||||
| Baseline to 3 months | 13/17 (76.5%) | 12/19 (63.2%) | 0.481 | ||||
| Baseline to 3 weeks | 7/17 (41.2%) | 4/19 (21.1%) | 0.281 | ||||
| 3 weeks to 3 months | 6/17 (35.3%) | 8/19 (42.1%) | 0.742 | ||||
RAGT: robot-assisted gait training, mRS: modified rankin scale. The values represent absolute and relative frequencies (%). The p value was obtained using the Fisher’s exact test
Table 3 showed changes in 10MWT, TUG, and BBS for the RAGT and conventional groups over 3 weeks and 3 months. As 10MWT and TUG require the ability to walk at least a few meters independently, corresponding to a FAC score of ≥3, these tests were feasible only in 8 RAGT and 10 conventional group patients at baseline. The RAGT group showed significant improvements in all measures at both intervals. In the conventional group, 10MWT improvement was not significant. TUG and BBS improvements were significantly greater in RAGT patients at both time points. No further significant 10MWT improvement occurred between 3 weeks and 3 months after RAGT ended.
Table 3.
Analysis of differences in walking ability parameters
| RAGT |
Conventional |
p b | |||||
|---|---|---|---|---|---|---|---|
| n | Improvement | p a | n | Improvement | p a | ||
| 10MWT improvement, m/sc | |||||||
| Baseline to 3 months | 8 | 0.24 (0.14; 0.70) | 0.008 | 10 | 0.36 (0.07; 0.55) | 0.065 | 0.897 |
| Baseline to 3 weeks | 8 | 0.26 (0.11; 0.53) | 0.008 | 10 | 0.24 (0.01; 0.41) | 0.080 | 0.829 |
| 3 weeks to 3 months | 8 | 0.05 (0.01; 0.19) | 0.055 | 10 | 0.07 (-0.01; 0.19) | 0.080 | >0.999 |
| TUG improvement, sd | |||||||
| Baseline to 3 months | 8 | 11.2 (10.4; 12.3) | 0.008 | 10 | 6.4 (4.6; 8.1) | <0.001 | 0.034 |
| Baseline to 3 weeks | 8 | 9.5 (7.7; 12.2) | 0.008 | 10 | 6.0 (4.0; 6.4) | <0.001 | 0.043 |
| 3 weeks to 3 months | 8 | 1.0 (−0.6; 2.1) | 0.098 | 10 | 0.6 (−0.5; 2.8) | 0.116 | 0.898 |
| Berg Balance Scale improvementc | |||||||
| Baseline to 3 months | 17 | 28 (19; 36) | <0.001 | 19 | 15 (10; 28) | <0.001 | 0.049 |
| Baseline to 3 weeks | 17 | 17 (10; 26) | <0.001 | 19 | 11 (6; 17) | <0.001 | 0.027 |
| 3 weeks to 3 months | 17 | 9 (2; 13) | <0.001 | 19 | 5 (2; 11) | <0.001 | 0.525 |
10MWT: 10-meter Walk Test, RAGT: robot-assisted gait training, TUG: timed up and go. The values represent the median and interquartile range. The p value was obtained with the paired Wilcoxon testa or the Mann–Whitney testb. Differences were defined according to the expectation of an increasec or decreased in a given parameter. The corresponding one-tailed paired Wilcoxon test was applied
Discussion
In our single-center randomized controlled trial targeting patients in the subacute phase of their first-ever ischemic stroke, we found that adding a 3-week protocol-defined RAGT offers advantages over conventional therapy alone in terms of ≥2-point FAC improvement, as well as TUG and BBS outcomes during the 3-month follow-up.
The importance of the initial 10–12 weeks following a stroke for the recovery of gait is widely acknowledged.[2] The FAC, being readily applicable in routine clinical settings, serves as the most widely employed, reliable, and validated scale for the objective assessment of functional gait recovery. All our patients exhibited an improvement of at least 1 point in their FAC scores over the course of 3 months, which is in congruence with the meta-analysis by Mehrholz et al.[9] Within the initial 3 weeks, such improvement was observed in all patients from the RAGT group, contrasting with only 78.9% of participants undergoing conventional treatment. Notably, all participants in our RAGT group exhibited improvements in their FAC scores within the initial 3 weeks, a trend not mirrored in the conventionally treated group (albeit without statistical significance). Likewise, the percentage of patients in the RAGT group showing improvements of 2, 3, or 4 points consistently surpassed that of the conventionally treated group at both the 3-week and 3-month intervals (although statistical significance was not attained). The observation of a consistent trend favoring superior outcomes in the RAGT group at both assessment time points and across various magnitudes of improvement suggests a potential benefit conferred by incorporating RAGT therapy in addition to conventional rehabilitation approaches. Interestingly, the group receiving RAGT appears to have continued improving in FACs more than the conventionally treated group even after discontinuation of RAGT treatment, with 66.7% of eligible RAGT patients improving by 2 points or more between 3 weeks and 3 months, compared to only 21.4% in the conventional rehabilitation group. Previous trials examining the effects of RAGT alongside conventional rehabilitation have suggested an increased probability of patients achieving independent walking within 6 months following a stroke in the RAGT cohort. However, these findings are based on low-quality evidence as per the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) criteria. Notably, most of these trials involved randomization well beyond the initial 14 days post-stroke, and they were characterized by notable methodological shortcomings. These deficiencies included issues related to randomization procedures, recruitment long after the index event, outcome assessment methods, and/or a lack of clear definitions of rehabilitation protocols. Additionally, the trials suffered from small study groups, which further limited their generalizability.[9,18]
The use of RAGT after stroke has demonstrated promising outcomes in terms of balance recovery, as evidenced by a recent meta-analysis by Baronchelli et al.[19]. In their meta-analysis, the TUG consistently indicated improvements in balance among individuals undergoing Lokomat treatment. Their conclusions regarding the BBS were, however, less definitive, with results varying and showing no clear trend toward improvement in BBS. This might suggest that while the addition of RAGT may positively impact certain aspects of balance recovery post-stroke, its effectiveness in addressing all facets of balance impairment remains uncertain.
In this trial, the addition of Lokomat ProFreeD to conventional treatment yielded superior outcomes (i.e., improvements relative to baseline) in both BBS and TUG scores at 3 weeks and 3 months. The improvement in 10MWT was significant in the RAGT group but not in the conventional group. However, as the trend toward improvement was relatively strong even in the conventionally treated group, the difference in improvements between the RAGT and conventional groups was insignificant. Notably, the progression observed after discontinuing RAGT, specifically within the 3-week to 3-month period, did not show significant differences in further post-rehabilitation improvement between the groups. This observation suggests that the superior outcomes in the RAGT group noted after 3 months (i.e., greater improvements in the RAGT group over the conventionally rehabilitated group relative to baseline) may largely stem from the initial 3-week period involving RAGT. Further investigation is warranted to elucidate the mechanisms underlying these observed improvements and to discern the sustained effects of RAGT on long-term balance recovery post-stroke.
The improvement in mRS scores did not differ significantly between the groups in our study, although the improvement was consistently higher in the RAGT group compared to the conventional group. Considering the observed trends, a higher number of patients would likely yield statistical significance in mRS as well. However, this study is the first to evaluate the effect of RAGT on mRS in subacute post-stroke patients and, as such, we had no way to estimate the required sample size with this parameter in mind (note that mRS was only one of the secondary endpoints in our study). Therefore, this study also provides an estimation of the population size necessary to detect a possible difference between RAGT and conventional groups in mRS.
Still, the relatively modest sample size can be considered a limitation. Unfortunately, the study’s enrollment and dropout rates were influenced by disruptions caused by the COVID-19 pandemic during the study period, contributing to challenges in participant recruitment and retention. Therefore, the study population did not reach the originally estimated size (50 per arm). Nevertheless, significant results were detected for the primary outcome (FAC) as well as for TUG and BBS parameters, indicating that for these parameters, the study had sufficient power. Moreover, our study has several strengths, including stratified randomization and the well-defined early post-stroke initiation of rehabilitation therapy, following the latest rehabilitation recommendations by the European Stroke Organization,[20] as well as a relatively wide range of endpoints evaluated by certified personnel. Additionally, the development and implementation of our well-defined rehabilitation protocol, utilizing Lokomat ProFreeD, were guided by current evidence outlined in the Cochrane meta-analysis.[9]
The pilot nature of the current study and the fact that it was conducted in a single specialized rehabilitation center with dedicated RAGT equipment and trained personnel may limit the generalizability of the findings. Additionally, the substantial exclusion rate (141 of 184 screened patients, 76.6%) resulting from reasonable inclusion and exclusion criteria (ensuring tolerability of RAGT by patients and, however, not rehabilitating patients who do not need this therapy) indicates that this therapy is likely to be beneficial only for a selected subgroup of stroke patients.
It is true that the tolerability of individual patients to the therapy and their physical condition differed, which precluded us from offering all patients the identical total therapy time. However, this is a general limitation across most rehabilitation studies. We have, however, tried to address this by predominantly using a paired design, i.e., primarily evaluating improvements of individual patients rather than the total performance of the group.
Conclusions
Our study explored the addition of protocol-defined RAGT to standard protocol-defined conventional rehabilitation, which is the prevailing standard of care in our institution. This combined approach, when initiated within 4 weeks after the occurrence of a first-ever ischemic stroke (median time to randomization was only 13 days), demonstrated positive effects on gait and balance (in particular, FAC, TUG, and BBS). Notably, the group subjected to RAGT displayed sustained improvements in FACs even after the cessation of RAGT treatment, suggesting a potential enduring impact that warrants further investigation in subsequent studies.
Author contributions
IF: conceptualization; project administration; funding acquisition. OV: conceptualization; methodology; writing—original draft (lead). SB: editing. TA: writing—review and editing. PH: methodology; AG: conceptualization (supporting); writing—review and editing. MB: validation; supervision. AK: data curation; statistical analysis.
Conflicts of interest
There are no conflicts of interest.
Data availability statement
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.
Acknowledgments
We thank all members of the interdisciplinary team for their support and all study participants for their cooperation. Our gratitude also extends to the technical staff for their indispensable assistance throughout the study.
Funding Statement
Supported by the institutional grant: MH CZ-DRO-FNOs/2020.
References
- 1.Mehrholz J, Wagner K, Rutte K, Meissner D, Pohl M. Predictive validity and responsiveness of the functional ambulation category in hemiparetic patients after stroke. Arch Phys Med Rehabil. 2007;88:1314–9. doi: 10.1016/j.apmr.2007.06.764. [DOI] [PubMed] [Google Scholar]
- 2.Winstein CJ, Stein J, Arena R, Bates B, Cherney LR, Cramer SC, et al. Guidelines for adult stroke rehabilitation and recovery a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2016;47:E98–169. doi: 10.1161/STR.0000000000000098. [DOI] [PubMed] [Google Scholar]
- 3.Harvey RL. Predictors of functional outcome following stroke. Phys Med Rehabil Clin N Am. 2015;26:583–98. doi: 10.1016/j.pmr.2015.07.002. [DOI] [PubMed] [Google Scholar]
- 4.Maier M, Ballester BR, Verschure PFMJ. Principles of neurorehabilitation after stroke based on motor learning and brain plasticity mechanisms. Front Syst Neurosci. 2019;13:74. doi: 10.3389/fnsys.2019.00074. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Stinear CM, Lang CE, Zeiler S, Byblow WD. Advances and challenges in stroke rehabilitation. Lancet Neurol. 2020;19:348–60. doi: 10.1016/S1474-4422(19)30415-6. [DOI] [PubMed] [Google Scholar]
- 6.Morone G, Paolucci S, Cherubini A, De Angelis D, Venturiero V, Coiro P, et al. Robot-assisted gait training for stroke patients: Current state of the art and perspectives of robotics. Neuropsychiatr Dis Treat. 2017;13:1303–11. doi: 10.2147/NDT.S114102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Calabró RS, Sorrentino G, Cassio A, Mazzoli D, Andrenelli E, Bizzarini E, et al. Robotic-assisted gait rehabilitation following stroke: A systematic review of current guidelines and practical clinical recommendations. Eur J Phys Rehabil Med. 2021;57:460–71. doi: 10.23736/S1973-9087.21.06887-8. [DOI] [PubMed] [Google Scholar]
- 8.Moucheboeuf G, Griffier R, Gasq D, Glize B, Bouyer L, Dehail P, et al. Effects of robotic gait training after stroke: A meta-analysis. Ann Phys Rehabil Med. 2020;63:518–34. doi: 10.1016/j.rehab.2020.02.008. [DOI] [PubMed] [Google Scholar]
- 9.Mehrholz J, Thomas S, Kugler J, Pohl M, Elsner B. Electromechanical‐assisted training for walking after stroke. Cochrane Database Syst Rev. 2020;5:CD006185. doi: 10.1002/14651858.CD006185.pub5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pollock A, Baer G, Campbell P, Choo PL, Forster A, Morris J, et al. Physical rehabilitation approaches for the recovery of function and mobility following stroke. Cochrane Database Syst Rev. 2014;2014:CD001920–CD. doi: 10.1002/14651858.CD001920.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kim YH. Robotic assisted rehabilitation therapy for enhancing gait and motor function after stroke. Precis Future Med. 2019;3:103–15. [Google Scholar]
- 12.Fiedorová I, Chmelová I, Hanzlíková P, Eliáš P, Šalounová D, Bar M, et al. Study protocol -robot-assisted gait therapy using Lokomat ProFreeD in patients in the subacute phase of ischemic stroke. Czech and Slovak Neurology and Neurosurgery. 2021;84:361–6. [Google Scholar]
- 13.Park IJ, Park J-H, Seong HY, You JH, Kim SJ, Min JH, et al. Comparative effects of different assistance force during robot-assisted gait training on locomotor functions in patients with subacute stroke: An assessor-blind, randomized controlled trial. Am J Phys Med Rehabil. 2019;98:58–64. doi: 10.1097/PHM.0000000000001027. [DOI] [PubMed] [Google Scholar]
- 14.Ochi M, Wada F, Saeki S, Hachisuka K. Gait training in subacute non-ambulatory stroke patients using a full weight-bearing gait-assistance robot: A prospective, randomized, open, blinded-endpoint trial. J Neurol Sci. 2015;353:130–6. doi: 10.1016/j.jns.2015.04.033. [DOI] [PubMed] [Google Scholar]
- 15.Watanabe H, Tanaka N, Inuta T, Saitou H, Yanagi H. Locomotion improvement using a hybrid assistive limb in recovery phase stroke patients: A randomized controlled pilot study. Arch Phys Med Rehab. 2014;95:2006–12. doi: 10.1016/j.apmr.2014.07.002. [DOI] [PubMed] [Google Scholar]
- 16.Banks JL, Marotta CA. Outcomes validity and reliability of the modified Rankin SCALE: Implications for stroke clinical trials: A literature review and synthesis. Stroke. 2007;38:1091–6. doi: 10.1161/01.STR.0000258355.23810.c6. [DOI] [PubMed] [Google Scholar]
- 17. Correct form: https://www.neuropt.org/practice-resources/anpt-clinical-practice-guidelines . [Google Scholar]
- 18.Tedla JS, Dixit S, Gular K, Abohashrh M. Robotic-assisted gait training effect on function and gait speed in subacute and chronic stroke population: A systematic review and meta-analysis of randomized controlled trials. Eur Neurol. 2019;81:103–11. doi: 10.1159/000500747. [DOI] [PubMed] [Google Scholar]
- 19.Baronchelli F, Zucchella C, Serrao M, Intiso D, Bartolo M. The effect of robotic assisted gait training with Lokomat® on balance control after stroke: Systematic review and meta-analysis. Front Neurol. 2021;12:661815. doi: 10.3389/fneur.2021.661815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kwakkel G, Stinear C, Essers B, Munoz-Novoa M, Branscheidt M, Cabanas-Valdés R, et al. Motor rehabilitation after stroke: European Stroke Organisation (ESO) consensus-based definition and guiding framework. Eur Stroke J. 2023;8:880–94. doi: 10.1177/23969873231191304. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.