How VNS-REHAB Aims To Improve Stroke Recovery After Patients Have Plateaued

A conversation with Brain Recovery Lab, Mass General Brigham University of Health Professions, Lead Investigator, and VNS-REHAB Trial Director Teresa Kimberley, PT, Ph.D., FAPTA, FASNR, and Clinical Leader Executive Editor Abby Proch

The potential for recovery plateau after ischemic stroke is a reality for many. The greatest gains occur in the weeks and months after stroke, while meaningful functional limitations may persist long term.

Now, that might not be the case.

Chronic ischemic stroke patients — those who are six months or more away from their stroke event — have the ability to regain function long after what was originally thought possible.

In this conversation, Teresa Kimberley, PT, Ph.D., FAPTA, FASNR, lead investigator and VNS-REHAB trial director at the Brain Recovery Lab, Mass General Brigham University of Health Professions, discusses what it took to evaluate the organization’s Vivistim Paired VNS System in combination with intensive outpatient rehabilitation therapy.

Kimberley explains why the pivotal study focused on patients at least nine months post-ischemic stroke, how the team standardized therapy across sites, and why participant expectations, therapist training, and adherence were central to the protocol’s success. She also reflects on the challenges that remain after FDA approval, including inconsistent insurance coverage for breakthrough treatments.

Clinical Leader: The inclusion/exclusion criteria specified that patients should be at least nine months out from their ischemic stroke. What purpose does that limitation serve?

Kimberley: The feasibility study used four months, and the pilot study used six months. The pivotal study used nine months to avoid any criticism that any further improvement might have occurred due to natural progression or natural improvement. The AHA Guidelines define chronic as six months, so the study used nine months to be conservative and avoid criticism.

Were there any recruitment challenges specific to enrolling patients with chronic ischemic stroke, especially when some may have been told their recovery had plateaued?

Indeed, people had been told they had plateaued, but they were motivated to try a new option. Recruitment challenges were primarily due to the surgical procedure; some patients considered participating but changed their minds after family or spouse expressed concern about the surgery. Others were comforted that the VNS surgical procedure had been approved for a long time. Another issue was the inability to use botulinum toxin for spasticity management. We could not allow botulinum toxin use as the effects are too variable and cyclical for study purposes, although it is helpful clinically. 

The trial studied the Vivistim Paired VNS System combined with intensive outpatient rehabilitation therapy. How did you account for differences between therapists’ approaches and attempt to standardize them within the protocol?

This is a very important point that we worked diligently on to have a high degree of fidelity in the treatment. The goal was to have a standardized therapy that could be individualized according to each participant’s level of impairment. Therapists were trained in the upper-limb rehabilitation protocol, including the specific order of tasks and the number of repetitions. Each task was to be performed with at least 30 to 50 repetitions for a total of at least 300 repetitions per session. Six to seven tasks were performed in each 90-minute session (Kimberley et al 2019). Study staff trained therapists in person and were present for the first few sessions when a therapist was treating someone for the first time to ensure consistency. We also videotaped the first sessions with each participant and reviewed them to give therapists feedback. Therapists were also encouraged to call us with questions, and we reviewed questions on the monthly full team call.

Not all patients benefited, as one would expect. How can you target patients who do respond more effectively? And will that necessitate another trial to do so?

A subgroup analysis did not show that any characteristic was associated with better response (e.g., younger patients or those with shorter time since stroke did not respond better) (Dawson 2023). A randomized trial large enough to show response subgroups is not planned due to time and expense.

However, the company is sponsoring a registry of commercially implanted patients, called the GRASP Registry. Currently, several hundred patients have been enrolled. We will analyze the registry data to determine any response factors. My lab is also interested in studying the physiology of VNS response to determine if we can optimize stimulation parameters.

How did you communicate the expectations and commitment involved to prospective participants, particularly around surgery, repeated therapy sessions, and the possibility of delayed or incremental gains?

Informed consent was obtained from all subjects, with a clear description of the time commitment, the possibility of nonresponse, and surgical side effects. Principal investigators and study coordinators needed training so they clearly understand the protocol and can answer any questions. I think it is important to assess prospective research participants and ensure their expectations are clear — one doesn’t want to enroll someone thinking they are going to receive a magical cure. Given the amount of work required from the stroke survivors for this protocol, one of my first questions to them was, “How did you like physical and occupational therapy?”

I feel it is always important to explain the rationale behind what you are asking of the participant. In this case, the Vivistim device enhances neuroplasticity — the brain’s ability to reorganize to make new beneficial connections. Intense physical and occupational therapy with a target of ~300 repetitions (which requires ~60 to 90 minutes) of challenging, meaningful movement is known to drive the brain’s natural neuroplastic processes and is the best practice therapy for stroke recovery. The project’s goal was to pair VNS stimulation with this best-practice therapy to maximize the beneficial brain effect. 

Unfortunately, due to reimbursement and outdated therapy methodologies, people with stroke often don’t get adequate therapy. In fact, there has been a decrease in available therapy for patients with stroke over the past decade. Participants were hungry for something else to help them. Protocols that offer treatment for people in the chronic stage of the disease do not struggle to enroll people.

Similarly, how did you approach patient engagement and adherence in your latest trial, given the device requires not only implantation but also rigorous physical therapy?

We had excellent adherence. Only one of the 108 patients from the pivotal trial (Dawson et al, 2021) did not complete the treatment (Vivistim plus rehab) protocol. Half of the patients actually completed double the regimen, since sham control patients crossed over after the three-month follow-up and repeated the therapy with the active Vivistim therapy.

A key strength of our study was training therapists and having them be part of the trial versus using standard of care therapy. Participants engaged with study staff repeatedly over the course of a year, and that relationship helps to solidify adherence. Other important factors include ensuring that participants are appropriately screened and understand what they are signing up for. With Vivistim, although patients must make the decision to undergo a surgical implant, once implanted, patient engagement and adherence is extremely high. We have shown this in our two-year follow up paper (Kimberley et al 2026).

What challenges did you face, if any? And how did you address them?

The biggest challenge hasn't been what people may expect. Patient engagement, adherence, and compliance have not been the limiting factors; patients are remarkably committed to the therapy.

The dominant challenge is securing insurance approval. Coverage for newly authorized devices remains slow and inconsistent, even for those carrying FDA Breakthrough Device designation. Research from Josh Makower's group at the Stanford Mussallem Center for Biodesign, published in JAMA Health Forum, found that the median time from FDA authorization to even nominal Medicare coverage was 5.7 years.

For stroke survivors, a delay of that magnitude could mean that they are unable to benefit from new technologies in their lifetime. Delays spent waiting for a coverage decision may mean patients will never receive a therapy that has already met the FDA's standard for safety and effectiveness.

It is difficult to justify that gap on clinical or ethical grounds. When a therapy has cleared the evidentiary bar the FDA sets, the burden should shift toward covering it rather than toward the patient and clinician having to prove the case again, patient by patient and payer by payer. CMS's newly detailed RAPID pathway, which would let the same clinical evidence used for FDA approval support a Medicare coverage decision, is a meaningful step, but the underlying problem is structural and still largely unaddressed by private payers.

It is immensely frustrating to see your research achieve the pinnacle of FDA approval, only to have so many hurdles continue.

Looking back, are there lessons from VNS-REHAB that could inform the design of future neurorehabilitation trials, especially when it comes to reducing burden while still capturing meaningful functional outcomes?

Our trial design used best practice methodologies (randomized, blinded, sham-controlled) and carefully considered the appropriate patient population. The patient selection was well informed by preclinical work and two pilot trials. Often, companies are inclined to skip these important steps.

One challenge for an implantable device is the blinding and sham-controlled aspect. Future neurorehabilitation trials should utilize triple-blinding methodology. That is, the patient, the treating therapist, and the assessor are all unaware of group designation. Critically, our study used different therapists to treat and assess, avoiding the possibility that the treating therapist might tend to score a patient higher.

Many neurorehabilitation trials do not use designs that are this rigorous. Unfortunately, that means it prevents them from providing persuasive evidence, even if they show positive results.

To reduce the burden, the study must be well designed to capture meaningful data and outcomes appropriate to the treatment. A less carefully considered design may be less costly or quicker to perform, but it ultimately adds to the overall burden because additional studies must be conducted to understand long-term effects or persuade clinicians of the therapy's usefulness. 

About The Expert:

Teresa Jacobson Kimberley, PT, Ph.D., FAPTA, FASNR is professor and director of the Brain Recovery Lab in the Department of Physical Therapy in the Mass General Brigham University of Health Professions in Boston, MA. She also serves as a research associate in the Department of Neurology at Mass General Hospital, Spaulding Rehabilitation Hospital, and Harvard Medical School. Her lab studies neuroplasticity and strives to understand the pathophysiology of motor impairment and develop novel rehabilitation interventions for neurologic disorders. She has over 15 years of experience leading clinical trials investigating novel neuromodulatory interventions in individuals with neurological diseases and has been continuously funded by the NIH, foundations, and industry for 20 years. She led the development of the rehabilitation protocol for Paired VNS and was one of the lead scientists on the pilot and pivotal Paired VNS REHAB trial.