The Evolution of Poststroke Recovery: Advances, Challenges

A seismic shift is underway in how experts view functional recovery after a stroke.

For years, the prevailing belief was that regaining limb mobility beyond 6 months was nearly impossible, as damaged brain circuits were thought to be irreparable. But today, a wave of renewed optimism — fueled by groundbreaking research — is challenging that notion. From brain stimulation to stem cell therapy, scientists are exploring new ways to push the boundaries of recovery.

It’s still early days, but the potential impact of these emerging interventions could be profound. Currently, approximately seven million Americans live with chronic stroke, with an estimated 75% experiencing some level of impairment.

As the population ages, stroke rehabilitation is poised to take center stage. By 2050, the number of stroke survivors in the United States is expected to nearly triple to 20 million, underscoring the urgent need for more effective recovery strategies.

Now, a growing body of research is challenging long-held beliefs about the limits of stroke recovery. Breakthroughs in neuroplasticity, brain stimulation, and regenerative medicine are opening new doors — offering fresh hope to patients and transforming stroke rehabilitation.

Brain Stimulation

Noninvasive brain stimulation — such as transcranial direct current stimulation (tDCS) or transcranial magnetic stimulation (TMS) — is being explored as a way to enhance stroke recovery by pairing physical rehabilitation with targeted brain stimulation to modulate disrupted neural networks.

However, key questions remain about the optimal dose, duration, and patient population. For example, findings from the TRANSPORT 2 trial, presented at the International Stroke Conference (ISC) 2025, revealed that tDCS, delivered at two different doses, did not improve upper extremity function beyond the gains achieved through physical therapy alone.

Despite the disappointing results from this relatively small phase 2 trial, the research is far from over. “This doesn’t close the door on tDCS,” said Lauren H. Sansing, MD, professor of neurology at Yale School of Medicine, New Haven, Connecticut, and chair of ISC. “It didn’t show a benefit, but it provided an enormous amount of data to help us rethink how to test this therapy.”

Joseph Broderick, MD, professor of neurology and rehabilitation medicine, University of Cincinnati in Cincinnati and director of the UC Gardner Neuroscience Institute, noted that determining the correct dose is essential as too much stimulation “can cause side effects that you don’t want.”

In addition, Wayne Feng, MD, professor of neurology and biomedical engineering, Duke University School of Medicine, Durham, North Carolina, who presented the TRANSPORT 2 results, acknowledged the lack of solid evidence for tDCS but insisted “it has an edge over TMS. [tDCS] is portable, easy to use, low cost, and can be paired with rehab therapy” which can’t be said for TMS, he said.

Another noninvasive brain stimulation technique, low-intensity transcranial focused ultrasound stimulation, targets cortical, subcortical, and deep brain regions.

While being explored for stroke recovery, it is also under investigation for a range of other neurological and psychiatric conditions.

Researchers are also exploring more direct methods of neuromodulation.

Deep brain stimulation (DBS) delivers constant electrical stimulation to a particular brain region through a surgically implanted thin wire or electrode.

One small trial targeting the dentate nucleus in the cerebellum — an area involved in coordinating muscle movements — showed that when combined with physical therapy, the approach improved upper limb function in stroke patients with persistent moderate to severe upper-extremity impairment. Participants showed a median seven-point improvement on the upper-extremity Fugl-Meyer scale.

Gary K. Steinberg, MD, PhD, Lacroute-Hearst professor of neurosurgery and neurosciences at Stanford University in California, believes electrical brain stimulation will play a key role in poststroke rehabilitation. Among the various approaches, he sees DBS as particularly promising.

However, some experts caution that DBS is relatively invasive, requires hospitalization, and carries potential neurosurgical risks.

Joel Stein, MD, professor and chair, Department of Rehabilitation and Regenerative Medicine, Columbia University, New York City, said he’s somewhat skeptical about DBS. “I have found it’s hard to convince patients to undergo a surgical intervention when the benefits may be modest,” he told Medscape Medical News. 

This hesitation around invasive procedures has led researchers to explore less invasive alternatives for stroke rehabilitation.

Vagus nerve stimulation (VNS) has been shown to be safe and effective in several neurological conditions, including epilepsy and stroke. Following a positive phase 3 sham-controlled trial in 2021, the US Food and Drug Administration approved a VNS device for patients in the chronic stage of ischemic stroke (> 9 months poststroke) with persistent moderate to severe upper extremity impairment.

Steinberg emphasized the benefit is modest and noted VNS is only for upper limb weakness, requires repeated stimulation, and has to be paired with intensive rehabilitation.“Nomatter what form of energy you use to stimulate the brain, whether it’s electrical or magnetic, the brain is going to be a better target than the vagal nerve,” he said.

Some experts argue that the precise mechanisms behind VNS and the consistency of stimulation parameters in stroke treatment require further investigation.

Exploring Stem Cell Therapy

This ongoing need for refinement in neuromodulation therapies has also driven interest in regenerative approaches, such as stem cell therapy.

Stem cell therapy involves injecting human neural stem cells through a small hole in the brain, a process designed to support neurorestoration. These stem cells secrete powerful growth factors, along with other proteins and molecules, that promote brain repair.

These factors help neurons grow new axons and dendrites, stimulate the formation of blood vessels in stroke-damaged areas, and reduce inflammation, all of which contribute to the recovery process.

The first in-human phase 1/2a study, conducted by Steinberg and colleagues, enrolled 18 patients at least 6 months poststroke but some years out from their stroke.

All patients showed improvement in their total Fugl-Meyer motor score at 12 months, with 10 out of 16 achieving clinically meaningful recovery — defined as an increase of 10 or more points on the scale.

Gains were observed in both upper and lower limb function. Upper limb improvement at 12 months (+6.9; P < .0002) exceeded that seen with VNS which showed an improvement of only 5.3 points, according to Steinberg. He emphasized that this was despite “our patients being much more severe” than those in the VNS trial.

The intervention also improved gait, quality of life, and even benefited patients with aphasia. Additionally, factors such as age, gender, and stroke volume “didn’t matter” in terms of the extent of improvement, Steinberg noted.

All adverse events, which included incisional pain, nausea, and fatigue, resolved spontaneously and none were related to the stem cells.

Steinberg and his colleagues are now planning a randomized multi-center, blinded study of stem cell therapy.

Broderick cautioned that the study is too small to determine whether patients would have improved without the intervention. He noted that many treatments have generated excitement early on, only to show diminished impact when tested in larger patient populations.

Stem cell therapy has also been explored in Parkinson’s disease, a condition that may be more suited to such an approach because it targets a specific population of cells, Broderick noted. However, even in that context, “it hasn’t worked out that well” and has been associated with side effects, he added.

This uncertainty also extends to stroke treatment, where many fundamental questions remain unanswered.

Feng emphasized the need to determine which cells should be injected or infused, in what quantities, and which stroke patients are most likely to benefit.

Robotics, Virtual Reality

As researchers continue to refine biological interventions like stem cell therapy, technological advancements are also playing a growing role in stroke rehabilitation.

This fast-growing field involves using robots or programmable devices designed to deliver high intensity task-specific training. Growing evidence suggests robotic therapy can boost limb function and motivate patients to stick to rehab exercises.

This is important as sustained repetitive exercises are crucial for poststroke recovery, but many patients drop out due to cognitive and other stroke-related issues.

However, Feng noted that research on robotic therapy for severely impaired patients has been less than encouraging, potentially highlighting the limits of brain plasticity. “As much as we have fantasies about brain plasticity, it’s quite limited in stroke patients once the brain has been ‘broken’ or injured by stroke,” he said.

Despite these challenges, researchers continue to explore alternative rehabilitation strategies. One emerging approach is innovative virtual reality (VR) therapy, along with interactive video gaming, which is gaining popularity as a neurorehabilitation modality.

A recent meta-review of 57 systematic reviews, encompassing 1033 randomized controlled trials, concluded that VR can improve upper and lower limb function, balance, gait, and possibly cognition.

VR offers several advantages over other rehabilitation approaches. It is relatively affordable, enhances patient satisfaction and engagement, and has no significant side effects.

“Exercise is a really critical piece of recovery, but it’s very hard to get people to comply, so if you can make it more engaging, that’s really important,” said Stein.

In addition, VR is relatively accessible, using, for example, video games offered by Nintendo Wii and Xbox Kinect. Some companies are attempting to make games more stroke-specific, said Stein.

Broderick pointed out that VR and other poststroke interventions do not address deficits in language and vision caused by stroke. “There are limits to what we can recover from,” he said.

A variety of compounds — including levodopa, fluoxetine, D-amphetamine, citicoline, niacin, and inosine — are being tested in animal models and clinical trials or are already in use to aid motor recovery after stroke. However, study results have been mixed.

Looking ahead, Broderick believes gene therapy could eventually help “spark motor recovery” in some stroke patients. He cited a study suggesting that certain genetic variants may influence stroke recovery outcomes.

Feng reported he is a consultant for NAMSA, and an advisor for Burke rehabilitation Institute.Stein reported he is on the advisory board of Dessintey, which develops and markets intensive rehabilitation technologies. He is also involved in a clinical research trial with Brain Q involving low intensity, noninvasive brain stimulation and his center is a participating site for a registry study of the vagal nerve stimulation device.

Steinberg’s stem cell work has received support from the California Institute for Regenerative Medicine. All other sources reported no relevant disclosures.