The results of this observational study provide evidence that a comprehensive and intensive eight-week rehabilitation program including BWSRGT followed by Manual Gait Training in patients early after stroke, can led to an improvement in all functional outcomes, independently of patient demographic or initial functional status. However, the results indicate that patients with initial FAC level of 2 or 3 obtained the most benefit. The intensive rehabilitation program was well tolerated, and no patients withdrew for factors related to the gait training or the high training dose.
Other research groups have studied the improvement of walking ability of sub-acute stroke patients combining methods of gait training showing that an intensive locomotor training on an electromechanical gait trainer plus physiotherapy resulted in a significantly better gait ability and daily living competence compared with physiotherapy alone[22, 31].
However, many studies have failed in showing significant differences in gain of functional scores when comparing robot-driven gait orthosis training with conventional physiotherapy. In the study by Peraula et al, chronic ambulatory patients regained the same walking ability when they received body weight supported training with or without FES compared with over-ground walking exercise training program.
The dose or intensity of the training seems to influence the improvement in the walking ability improvement since our study shows greater results than studies with only 20 or 30 min of daily therapy for 3 to 4 weeks[32, 33]. Higher intensity of gait practice, in line with modern principles of motor learning, probably explains the superior results. The total amount of rehabilitation given to the patients is higher than reported in other studies[22, 34] and our results should be considered in the context of high-intensity rehabilitation in sub-acute stroke.
In our study, after 8-weeks of intensive rehabilitation we found gait improvements in one or more of the outcome measures in 95.54% patients. This finding is higher than recovery reported by other authors[23, 35] and may indicate that the higher dose in our rehabilitation program can lead to greater improvement of motor function in sub-acute stroke patients. To determine the magnitude of the improvement attributed to gait training, a comparison with an experimental control group without gait training would need to be done. This considered, the results of the present study should be used as a benchmark for expected change to aid clinical decision-making and to power controlled clinical research studies.
The selected functional outcome measures were sensitive to detect change across patients and may be suitable to use for future studies. Care should be taken interpreting functional scales that may include the use of assistive technology (as can be used in the 10 MWT). The underlying factors of patient performance leading to improved scores on each outcome measures is difficult to determine from the present study. For example, the improved score on the Tinetti balance test could be a cause of improved gait, since various balance functions are known to affect gait[36, 37]. The more the patient sways, the worse is the balance and consequently the gait parameters. According to Kollen et al the recovery of independent gait is highly dependent on improvements in control of standing balance. These results are in line with our study, where the gait speed and functional ambulatory measures improve in parallel with balance measures.
A pertinent finding of our study is that patients with mid-range FAC at admission obtained the most benefit, which raises the possibility that FAC could be tested as a clinical predictor for recovery, although Masiero et al did not find a correlation between FAC and motor recovery during conventional rehabilitation programs. Other studies have shown that initial level of paresis or trunk control could be used as clinical predictors of balance and gait for rehabilitation in sub-acute patients. Moreover, previous studies have reinforced the idea of the excellent reliability of the FAC, good predictive validity and responsiveness in sub-acute stroke patients and it has been proposed to predict community ambulation with high sensitivity and specificity. We provided evidence that suggests that FAC could be useful as a predictor of outcome, but not initial walking speed, as reported by Barbeau.
One of the limitations of the present study is that the design of the study does not allow the comparison of the robotic gait training with the same amount - 60 min - of conventional gait therapy, combined both with an intensive rehabilitation program. The characteristics of the patients that would benefit the most of this type of combined gait therapies (robotic + conventional) remain unclear. In our study, patients ranged from the early phase of recovery to 3 months after the injury, when the largest gains are observable,[43–45] however some studies have found improvements in gait function in late phases of recovery. The optimum dose of therapy is another open question for future studies. Even if daily therapy seems to be a decisive factor in the training program success early after stroke, [14, 47] there is disparity of results about what is the ideal frequency of training in chronic phases[48, 49].