Important questions raised by the AMBULATE trial: a trial of treadmill and overground walking for stroke


By Richard Lindley

We then decided to explore the value of treadmill training and overground walking as an intervention to improve walking in the more chronic post-stroke phase. Following the favourable results for treadmill training with body weight support (MOBILISE Trial, Stroke 2010; 41: 1237-42) in helping people to walk early after stroke, we wanted to test the hypothesis that treadmill training with overground walking without body weight support practice could improve walking in those with chronic stroke disability. We hypothesised that if people could improve their walking speed to near normal, this would become self-sustaining as they would be more likely to continue to walk and thus retain this improvement. Normal walking speed for older people is 1.3m/s and at this speed, normal older men would cover 576m in six minutes (494m for women) but typical speeds for those with disability after stroke is about 0.4 to 0.8m/s.


The AMBULATE trial was a single-blinded randomised controlled trial with randomisation (stratified by baseline walking speed) with two intervention groups and a no-treatment control. The control group was merely assessed at baseline and at the routine follow-up points for the trial. One intervention group received two months of treadmill training and the second intervention group received four months of training. The programme was carried out in a community setting, and the participants were recruited about 20 months after their stroke.

They were all walking independently (and thus, body weight support was not used), but some required walking aids. Transport was provided when necessary. All participants received individualised programmes to increase step length, speed, workload, and automaticity by the same therapist. To increase step length, the treadmill was run at a comfortable speed and participants were instructed to “walk as slowly as possible” and/or a metronome was used to decrease cadence thereby encouraging larger steps. When necessary, marching-type steps were included to encourage hip and knee flexion during swing phase to improve toe clearance. When a normal step length was observed, the therapist increased the speed of the treadmill until step length was compromised. Overground walking was used in each session to reinforce gains achieved during treadmill training. Overground walking comprised 20% of intervention time in week one and was progressively increased each week so that it comprised 50% of the 30-minute intervention time in week eight of training. In week nine, the four-month training group returned to 20% overground walking which was again increased to 50% by week 16. Our outcome measures were collected at two months, four months, six months and 12 months after the baseline measures were taken. Outcome measures were collected by therapists trained in the measurement procedures who were blind to group allocation.

Our results showed that this training regime improved walking for both intervention groups, but unfortunately the improvements were not maintained. Four months of training was statistically better than two months of training so by the four-month follow-up, the four month training group walked 38m (95% CI 15 to 60) more than the control group and 29m (95% CI 4 to 53) more than the two-month training group. Walking speeds increased, from a baseline of 0.5m/s to 0.68m/s in the four-month training group, but again these speeds fell back to near baseline after training ceased.

What does this mean for those with stroke? Our conclusions, backed by many anecdotes from stroke survivors, is that training (and in this case training supervised by a trained therapist) using a mixture of treadmill and overground walking practice can improve walking but this improvement is not sustained unless the training regime is continued. Stroke survivors are therefore no different to those without stroke: to maintain fitness continued training is required. For those with walking limitation after stroke, training should probably continue indefinitely.

These results are consistent with the observation that older frail people with chronic disability often require bursts of rehabilitation to maintain abilities. Our results provide important new insights into services for those with stroke.

The traditional post-acute stroke rehabilitation model is clearly insufficient. We need to be smarter in designing more sustainable community training. However, we wonder how many fitness clubs and gyms would welcome those with stroke in attending their centre?

Like all good research, our results have spurred us on to explore other methods of achieving increased physical activity. The Australian National Heart Foundation runs a community programme of physical activity called “Heartmoves” and we are exploring new research to investigate whether this programme could be adapted for those with stroke. Physical activity remains the most promising intervention for healthy ageing—our challenge is to make this achievable for the many.

The full results of the AMBULATE trial are in-press in the International Journal of Stroke.

Richard I Lindley is the Moran Foundation professor of Geriatric Medicine, The University of Sydney, Sydney, Australia. Louise Ada from the University of Sydney and Cath Dean from Macquarie University were the other chief investigators