Subscribe to eTOC

Post-Stroke, At-Home and Treadmill Therapy in Rehabilitation Centers Found Equally Effective



, seated, and research assistant, left, use a body-weight treadmill system to help train a stroke patient with walking at the Brooks Rehabilitation Center in Jacksonville, FL.

Stroke patients who did home-based physical therapy at two months and six months after the initial insult recovered as well as those who did body-weight supported treadmill training.

Walking recovery programs that use body-weight-supported treadmills are used in rehabilitation centers around the country offering stroke patients a means of improving functional walking. But the largest study designed to test whether it works better than a more traditional home-based physical therapy program suggests that about half of stroke patients who began a program two months after the initial insult improved using either technique. Moreover, patients who began walking recovery programs using the body-weight system six months post-stroke also showed the same gains as those who began treatment four months earlier.

The study, which was funded by the NINDS and published in the May 26 New England Journal of Medicine, answers a critically important question about the use of the body-weight-supported treadmills. Improvement in the ability to walk is a major goal in physical therapy for stroke patients, said the lead study author Pamela Duncan, PT, PhD, professor in the Division of Physical Therapy at Duke University.

Dr. Duncan and her colleagues enrolled 408 people who had a stroke two months earlier and were considered to be moderately to severely impaired in their walking ability. Moderate impairment was measured by the person's ability to walk 0.4 to < 0.8 meters per second and severe impairment, at < 0.4 meters per second.

The locomotor training included stepping on a treadmill with partial body weight support and assistance by a therapist for 20 to 30 minutes a session followed by walking over ground for 15 minutes. The home-based intervention included very structured and progressive exercises that enhanced flexibility, range of motion in joints, strength in arms and legs, coordination and balance. The patients were seen at home three days a week but encouraged to walk every day.

Participants were randomized to three groups: training at two-months post-stroke on a treadmill with their body-weight supported in a harness as therapists helped them step along the treadmill (early locomotor training); identical training that occurred at six months post-stroke (late locomotor training); and a progressive exercise program to improve strength and balance with a physical therapist in the home that began two months post-stroke. All of the interventions included 36 sessions that ran for 90 minutes over 12 to 16 weeks. At one year, they assessed the study volunteers on improvements in walking speed, motor recovery, balance, functional status and quality of life.

Among their findings, the investigators reported that 52 percent improved — whether they received the treadmill training or the physical therapy at home, whether they began locomotor training at six-months or two months. (The odds ratio was 0.83 comparing early locomotor training with home exercise, and 1.19 comparing those in the late locomotor training group with the home exercise.)

The implications of the finding are enormous, said Dr. Duncan. Home exercise with a physical therapist is very feasible, she noted, and patients are more likely to comply with the intervention. “Even if you intervene late, the patients continue to recover,” said Dr. Duncan.

“We want to take these results and generalize them for all patients,” Dr. Duncan said. She is hoping that more home care agencies embrace the findings and that insurance companies pay for similar home-based interventions.


This study showed that both body-weight support exercise and physical therapy floor exercises designed for strength and flexibility worked equally well, and in fact, a group of patients on a waiting list who started the home therapy fared just as well as the others in the study at assessment at one year, said Joel Stein, MD, the Simon Baruch Professor and Chair of the department of Rehabilitation and Regenerative Medicine at Columbia University.

But, he noted, the study was not designed to determine how effective physical therapy is compared with no intervention. “There is the suggestion (based on the six-month data) that both interventions at least accelerated recovery, and perhaps that either one is more effective than usual care,” said Dr. Stein. “It is important to recognize that 82 percent of the participants received physical therapy as part of their ‘usual care’ during the study period; it is therefore not really possible to determine if physical therapy is effective based on this particular study.”

Jon Erik Ween, MD, director of the Stroke and Cognition Clinic in the Brain Health Centre Clinics at Baycrest in Canada, said that the Duke study is “really a study of the added benefit of focused “locomotor training” in addition to standard care in a small proportion (10 percent) of a population of stroke patients who all have finished acute rehabilitation and are living in the community.”

Dr. Ween said that the study does not provide details about what their stroke is really like beyond a coarse measure of stroke type, that is, large vessel, lacune or hemorrhage. “The size and location of the stroke in the neuraxis, and hence the impact on locomotor pattern generation is not specified,” he added.

“This kind of research needs to match treatment with deficit and time post onset,” Dr. Ween said. “Walking is a complex behavior with multiple determinants, as the authors point out. In addition, neural plasticity is also dynamic and changes over the course of time after a stroke depending on the type, location, and severity of each stroke. Suspended treadmill studies should be designed to target patients with impaired locomotor patterns at the appropriate time post-stroke when the locomotor pattern generators are most amenable to modulation.”

He said that the group designed to wait for six months can't really be considered a classic “placebo” control as 80 percent received ‘standard care,’ mostly front-loaded in the first six months. “The study does not specify what ‘standard care’ is, but surely it involves interventions on transfers and walking,” Dr. Ween said. But researchers agree that it would be unethical not to treat post-stroke patients, “particularly since there is plenty of evidence that doing something is always better than doing nothing. But, we could compare treatments in a cross-over design.”

He suggested that future studies on suspended treadmill training should begin earlier, perhaps in the weeks or month post-stroke. And he thinks that fMRI should be part of the research equation “both to describe and stratify subjects based on neural activity and to monitor response to intervention.”

Dr. Stein added that that findings at six months suggest that more therapy in the subacute period after stroke may accelerate recovery, a finding that needs further exploration, he said, but which provides additional encouragement to stroke patients to pursue rehabilitation early and often.

“There is a “critical need for studies examining the dose-response relationship between physical therapy (and other rehabilitation interventions) and outcome. We don't really know the right amount of therapy, nor the optimal timing. Unfortunately, much of current practice is determined by customer and insurance coverage, rather than being data driven.”

Steven Cramer, MD, professor of neurology, anatomy and neurobiology at the University of California-Irvine, agreed. He added that another key finding from the current study emerged regarding the delay of treatment. “The previous wisdom was that patients would plateau by three months post-stroke. But that is not the case at all. You can wait six months and still see remarkable gain. People with motor deficits have a solid chance of achieving useful gains with a course of well-designed physical therapy at home. There is a lot of potential to promote plasticity.”


The June issue of Continuum: Lifelong Learning in Neurology®, Neurorehabilitation, provides articles and videos about recovery following brain damage while offering the opportunity to earn up to 12 AMA PRA Category 1 Credits™. This issue is available now online at

“This issue highlights exciting new and emerging neurorehabilitative therapies, ongoing clinical research, and important clinical aspects of neurorehabilitation that have the potential to improve our patients' lives,” explains guest editor Mary L. Dombovy, MD, vice president of clinical neurosciences at Unity Health System and clinical associate professor at the University of Rochester.

This issue also features a CD of video material that complements the issue, illustrating aspects of neurorehabilitation including the impact of specific deficits on functional ability, the use of various therapy techniques, and the use of specialized equipment and technological devices.


Duncan PW, Sullivan KJ, Hayden SK, et al, for the LEAPS Investigative Team. Body-weight-supported treadmill rehabilitation after stroke. N Engl J Med 2011; 364:2026–2036.