, suggests new research backed by imaging data.
In secondary analysis of a previous study, training enabled both “good” and “limited” walkers to increase travel distance during a 2-minute walk. However, for “dual-task” walking, good walkers improved their distance by approximately 10 m after training, whereas limited walkers improved by only 1 m. Brain imaging showed increased brain activity in the limited walkers, which could reduce cognitive resources available for performing a second task while walking.
These findings, which were published online May 30 in Clinical Rehabilitation, may explain the apparent lack of superiority, shown previously, of dual-task training compared with single-task training for patients with stroke and impaired walking ability, researchers noted.
“Imaging data were consistent with our hypothesis that walking automaticity might explain these results,” said lead author, senior clinical research fellow at Oxford Brookes University, United Kingdom.
At baseline, participants who walked slowly had increased resting state connectivity between contralesional M1 and cortical areas associated with conscious gait control.
“In response to the intervention, we found increased connectivity with the precuneus in those who walked slowly at baseline, an adaptation that might support walking in more complex situations,” Dr. Collett said.
After stroke, many patients have difficulty walking while performing a second task, such as holding a conversation. Training in dual-task walking has provided uncertain benefits, according to clinical research.
In healthy individuals, walking is believed to be a largely automatic process that requires minimal executive resources. Previous studies have suggested that a certain minimum walking speed is required to enable automatic control of walking in the brain.
“We know that those with better walking ability after stroke are better able to cope with additional cognitive loads while walking,” said Dr. Collett. “Here, we proposed that increased automatic gait control may provide a mechanism whereby executive resources are freed up to attend to additional tasks,” he added.
The investigators further hypothesized that greater walking speed is required for automatic gait control. To test these hypotheses, they analyzed data from a previously conducted randomized trial of single- and dual-task walking interventions.
Trial participants were aged 18 years or older, had survived a stroke that had occurred at least 6 months before enrollment, had reduced 2-minute walk distance relative to their peers, and had no comorbid neurologic or psychologic disorders.
Over 10 weeks, participants underwent 20 sessions that included 30 minutes of walking on a treadmill. They were randomly assigned to undergo single-task walking or dual-task walking. The latter incorporated cognitive tasks as distractions.
Good versus limited walkers
In the current study, investigators analyzed various assessments that had been conducted at baseline and after completion of the training sessions, including distance on 2-minute walks with and without a distracting task. In addition, participants underwent imaging with functional near-infrared spectroscopy (fNIRS) and fMRI.
Using previous research as a basis, the researchers defined good walking speed as 0.8 m/sec. They categorized all participants, regardless of their intervention assignments, as having good walking capacity (0.8 m/sec or more) or limited walking capacity (less than 0.79 m/sec).
A total of 50 participants enrolled in the study (mean age, 62 years), and 45 completed the interventions. Of those who completed the interventions, 22 were randomly assigned to undergo single-task training, and 23 were assigned to dual-task training.
The researchers categorized 21 participants as having good walking capacity and 24 as having limited walking capacity. Participants in each category were divided approximately evenly between treatment assignments.
Barthel index score, which assesses functional independence, was higher in the group of good walkers.
Increased travel distance
Results showed that after the interventions, distance traveled during the single-task 2-minute walk increased by 8.9 m for good walkers and by 5.3 m for limited walkers. For the dual-task 2-minute walk, the distance traveled increased by 10.4 m among good walkers and by 1.3 m for limited walkers. Change from baseline on the dual-task walk was not significant for limited walkers.
There was no significant difference between good walkers and limited walkers in their perceptions of participation in community walking. Neither group increased its walking activity significantly following the interventions.
At baseline, limited walkers, in comparison with good walkers, had significantly greater activation in the contralesional hemisphere during dual-task walking, which consisted of incorporating a planning task.
In contrast, for many good walkers, there was a decrease in activation during dual-task walking. Activation in the contralesional hemisphere correlated negatively with dual-task 2-minute walk distance.
The researchers also found a negative correlation between activation and dual-task 2-minute walk distance when the second task was the Stroop task.
“The original trial was never designed or powered to compare groups formed by walking speed or test our automaticity hypothesis, and the results need to be viewed within this context,” said Dr. Collett. The small sample size did not allow the researchers to detect small effects of the intervention, especially in the imaging data, he added.
It also prevented the investigators from comparing limited walking and good walking groups according to whether they underwent the single-task or dual-task intervention, “which would be a superior way to investigate our hypotheses,” Dr. Collett said.
“The result of this study should be seen as exploratory, with further investigation needed,” he noted.
Helping stroke survivors to walk in the community is challenging, and new interventions that enable them to navigate complex surroundings need to be designed, said Dr. Collett. “Research is required to better understand the conscious and automatic contribution to gait control, especially with neurological impairment,” he added.
Overall, “our results suggest that improving automatic walking may be an initial step to improve capacity to respond to more complex walking interventions. However, [future] trials are required to test this,” he concluded.
The next frontier?
Commenting on the findings,, professor of neurology at Harvard University and senior neurologist at Beth Israel Deaconess Medical Center, Boston, said that “recovery and rehab are going to be the next frontier in stroke neurology, because there has to be a limitation in the present emphasis on acute care.”
Some patients do not receive acute care on time, and current treatment is not curative, added Dr. Caplan, who was not involved with the research.
Little scientific attention has been paid to how doctors can enhance recovery after stroke, what interventions delay recovery, and what the natural history of recovery is, he said. “This is a very nice study about that.”
Although the study’s methodology was sound, there were some limitations, including that strokes and underlying brain lesions were heterogeneous and that the study population was relatively small, Dr. Caplan said.
He added that “it’s a difficult study to do” and that it is difficult to organize participants into homogeneous groups.
Another limitation cited was lack of long-term follow-up that could indicate whether training provided sustained improvements in walking.
“It would be nice to revisit the same people later and see if their walking has improved, if they’re doing it differently, and if their subjective responses are different,” said Dr. Caplan.
In addition, the study did not examine whether the interventions made it easier for participants to walk with other people or to socialize more. “It may be that it really requires some time for them to gain confidence and for them to integrate that into their social network,” Dr. Caplan said.
“I would call it a proof-of-principle study, not a final study,” he noted. “It’s a study that shows that you can scientifically study rehab” and indicates the possible methodology that could be used.
The study was funded by the Stroke Association. Dr. Collett and Dr. Caplan have reported no relevant financial relationships.
A version of this article first appeared on.