Quick Overview
A Japanese hospital team ran a peer-reviewed randomised controlled trial of NMES, a form of EMS , after acute stroke.
They measured quadriceps muscle thickness by ultrasound on day 1 and day 14.
The weak thigh lost 2.65 mm with NMES versus 4.64 mm without it.
In milder stroke (NIHSS under 12), that thigh stayed at 0.00 mm versus a 3.88 mm drop.
Ankle tone was also lower after NMES.
The full group result was not statistically firm, so this is a signal, not proof.
The work came from Steel Memorial Yawata Hospital and is open access. The full post gives the protocol, safety notes, and limits.
We always provide direct links to the original research at the end of every article so you can review the evidence yourself.
EMS After Stroke: What a New Japanese Hospital Trial Really Measured
Stroke can steal strength almost overnight.
Muscle can shrink in the first days.
Researchers in Japan asked a simple question.
Could EMS, also called NMES, slow that loss?
This post explains that study.
It is not medical advice.
It does not sell a device.
It reports a published hospital trial.
Why this paper is a real study
The work appeared in Progress in Rehabilitation Medicine in 2025.
That journal is published by the Japanese Association of Rehabilitation Medicine.
Japan has a long record in hospital rehabilitation science.
The trial ran at Steel Memorial Yawata Hospital in Kitakyushu.
Partners included Kyushu University Hospital and the University of Occupational and Environmental Health Japan.
Lead author Kota Yamauchi is a registered physical therapist.
The hospital ethics committee approved the protocol.
Patients or families gave written consent.
Assessors who measured muscle did not know the group.
That is how a careful randomised controlled trial is built.
The problem they targeted
After acute stroke, the thigh can waste fast.
Doctors call this stroke-related sarcopenia.
It can reach about 42% of stroke patients in published reviews.
Loss of quadriceps muscle is linked with weaker walking and poorer function.
Muscle change can start within hours of a brain infarct.
The team wanted an add-on that works in bed.
Neuromuscular electrical stimulation can make a muscle contract without a full voluntary effort.
What they did
Sixty-three adults with a first one-sided stroke entered the trial.
All had an NIHSS score of 4 or more.
All arrived within 24 hours of onset.
Thirty-one were assigned to NMES.
Thirty-two were assigned to a very weak sham current.
Both groups received usual rehabilitation.
That meant 40 to 60 minutes of therapy most days.
The EMS group used a belt system for 30 minutes a day.
They completed 10 sessions over 2 weeks.
Settings were 20 Hz, a 250 μs pulse, and a 5:2 on-off cycle.
Staff raised the current until a contraction was visible.
Most patients tolerated about 4 to 6 mA.
The sham group felt only 0.1 mA.
That current did not make the muscle twitch.
What they measured
The main test was quadriceps muscle thickness on ultrasound.
Staff scanned the thigh on day one and day 14.
They measured both the weak side and the stronger side.
They also tracked strength, tone, trunk control, and walking category.
At three months they recorded the modified Rankin Scale.
Reliability checks were strong.
Two blinded testers had an inter-rater coefficient of 0.94.
Intra-rater reliability reached 0.99.
The numbers
On the weak thigh, thickness fell 2.65 mm with NMES.
It fell 4.64 mm without it.
The P value was 0.119.
That is not a statistically clear win for the whole group.
On the stronger thigh, the drop was 0.42 mm with NMES.
It was 1.93 mm in controls.
The P value was 0.223.
The effect size on the weak side was Cohen’s d = 0.53.
The trend favoured stimulation.
The trial was short of its planned size.
COVID-19 cut recruitment.
Power to spot a 6 mm gap fell to about 53%.
The authors say the study was underpowered.
People who stayed on NMES finished every planned session.
That is 100% adherence.
No treatment-related adverse events were reported.
The brighter signal
The team then split patients by stroke severity.
In milder cases, NIHSS under 12, the picture changed.
On the weak thigh, the EMS group showed 0.00 mm change.
Controls lost 3.88 mm.
That difference reached P = 0.043.
Only 23 people sat in that exploratory slice.
The authors call it hypothesis-generating.
It is not proof.
It is a clue worth a bigger trial.
Ankle stiffness was lower after two weeks of NMES.
The modified Ashworth score at the ankle favoured stimulation, P = 0.032.
Walking scores and three-month disability did not clearly differ.
Both groups had a median mRS of 4 at three months.
Why “more” matters in this science
The authors note that NMES effects often rise with dose.
This protocol used 30 minutes at 20 Hz.
Other stroke studies used 50 to 60 minutes.
Some used 50 Hz.
So the bright reading is not “a tiny buzz is enough.”
It is that a tolerated, visible contraction was safe here.
A fuller dose may be the next test.
More session time, more days, and a stronger protocol are the open door.
What this means for readers
The purpose of the trial was clear.
Can early EMS protect thigh muscle after acute stroke?
Across all patients, the answer was not yet statistically firm.
In milder stroke, muscle loss on the weak side looked smaller.
Tone at the ankle looked calmer.
The method was simple enough for a hospital ward.
This is hospital research, not a product pitch.
EMS here was an add-on to standard therapy.
It was not a replacement for walking practice or nursing care.
Anyone after stroke should follow their own clinical team.
The paper is open access under a Creative Commons licence.
The data sit in a peer-reviewed Japanese rehabilitation journal.
That is why this summary can be checked.
The headline finding is cautious.
The hopeful part is specific.
In people with less severe stroke, two weeks of NMES lined up with less quadriceps wasting on the affected side.
That is the story the measurements tell.
Not a miracle.
A signal.
And a reason better-powered trials should keep going.
Featured Product
ORIEMS FIT ULTIMATE KIT 💪 is a 100% Australian award-winning product.
Voted YEAR’S BEST ⭐ for 2024 and 2025.
It is a wide-range targeted muscle stimulator enhanced with EMS technology, designed to support your fitness and relaxation routine. 🌟
Trusted by more than 10,000 happy users with over 450 five-star Google reviews. 😊
Check availability.
Product Disclaimer:
This product is designed only to support fitness and relaxation routines. It is not a medical device and has not been evaluated or registered by the TGA. It is not intended to diagnose, treat, cure or prevent any disease or medical condition. It may not be suitable for everyone. Please consult your doctor or healthcare practitioner before using it.
Like this Research Digest? 📚
Share it with your friends 💬👇https://bit.ly/3USBLZJ
About Our Health Research Digest
We bring you the latest and most important health research papers — simplified and easy to understand.
Each study is broken down into clear, concise summaries and engaging podcasts, with direct links to the original research so you can explore further whenever you like.
We also create content based on your requests. If you’re curious about a specific health topic or would like us to find relevant studies, simplify the findings, and turn them into a podcast, just let us know.
Don’t hesitate to reach out — we’re here to help make health research accessible and useful for you. Contact us anytime.
More EMS Research Scientists Are Studying
1. Can EMS reduce fat?
2. Can EMS increase calorie burn while sitting?
3. Can EMS support muscle toning and muscle gains?
4. Can EMS improve athletic training?
5. Can EMS play a role in muscle loss & frailty?
6. Can EMS reduce pain?
7. EMS vs TENS: What are the differences?
8. Can EMS play a role after stroke?
Research Summary
| Topic | What the paper reports | Extra detail |
|---|---|---|
| Full paper title | Effect of Neuromuscular Electrical Stimulation on Muscle Atrophy in Patients with Acute Stroke: A Randomized Controlled Trial | Open-access article ID 20250034 |
| Authors | Kota Yamauchi, Yuri Hayashi, Shota Tanaka, Yoshihisa Fujino, Kenichi Kumagae, Kei Goto, Eisei Harayama, Shuji Arakawa | Lead author is a registered physical therapist |
| Journal and publisher | Progress in Rehabilitation Medicine, Volume 10 | Official journal of the Japanese Association of Rehabilitation Medicine |
| Where the work was done | Steel Memorial Yawata Hospital, Kitakyushu, Japan | Partners: Kyushu University Hospital; University of Occupational and Environmental Health Japan |
| Study design | Parallel two-group placebo-controlled randomised trial | Outcome assessors were blinded; full double-blinding was not possible because muscle twitches are visible |
| Study period and ethics | Enrolment July 2017 to June 2020 | Ethics approval No. 17-01; written consent from patients or families |
| Who was included | First-ever one-sided stroke, admitted within 24 hours, NIHSS score of 4 or higher | Cerebral infarction or haemorrhage confirmed on CT or MRI |
| How many people | 63 patients randomised; 31 to NMES and 32 to control | 24 NMES and 26 control completed the 14-day muscle analysis |
| What EMS protocol was used | Belt-type NMES (G-TES, Homer Ion, Tokyo) for 30 minutes a day | 10 sessions over 2 weeks; 20 Hz; 250 μs pulse; 5:2 on/off cycle; intensity usually 4–6 mA until a visible contraction |
| What the control received | Minimal current of 0.1 mA | No muscle contraction; some skin sensation only |
| Usual care in both groups | 40–60 minutes of task-oriented physiotherapy most days | Nutrition and other medical care were the same in both arms |
| Primary measurement | Change in quadriceps thickness on ultrasound from baseline to day 14 | B-mode scan of rectus femoris plus vastus intermedius; inter-rater ICC 0.94; intra-rater ICC 0.99 |
| Paretic-side result | NMES −2.65 ± 4.24 mm versus control −4.64 ± 4.58 mm | P = 0.119; Cohen’s d = 0.528; trend favoured NMES but was not statistically significant |
| Non-paretic-side result | NMES −0.42 ± 5.19 mm versus control −1.93 ± 3.36 mm | P = 0.223; Cohen’s d = 0.372 |
| Milder-stroke subgroup | NIHSS score under 12, n = 23 | Paretic side: NMES 0.00 ± 3.21 mm versus control −3.88 ± 5.11 mm; P = 0.043; authors call this exploratory |
| Other measured change | Ankle modified Ashworth score was lower after NMES | P = 0.032 at week 2; Motricity Index, FAC, TCT, and 3-month mRS did not differ clearly |
| Safety and adherence | Patients who continued NMES completed all 10 planned sessions | 100% session adherence in completers; no adverse events reported from stimulation |
| Main limit of the trial | Planned sample was 48 per group; COVID-19 cut recruitment | Detection power for a 6 mm difference fell to about 53%; authors call the trial underpowered |
| Authors’ conclusion | Two weeks of this NMES dose did not prove a preventive effect in the full sample | A possible benefit in milder stroke needs a larger trial; dose may have been modest versus 50–60 minute or 50 Hz protocols in other studies |
| Link to the original study | Official journal page and DOI | https://doi.org/10.2490/prm.20250034 · https://www.jstage.jst.go.jp/article/prm/10/0/10_20250034/_article · PubMed: https://pubmed.ncbi.nlm.nih.gov/41267677/ |
Related Posts Worth Reading
-
2026 Study coauthored by John Hopskin University School of Medicine , Published in Stroke / American Heart Association journal article
-
2015 Study - Published by BioMed Research International
-
A 2017 systematic review in Systematic Reviews, published by BioMed Central
-
2025 study published in Frontiers in Neurology by Frontiers Media, Switzerland.
-
2018 study published in Archives of Physical Medicine and Rehabilitation by Elsevier, Netherlands.
-
2024 study published in Archives of Physical Medicine and Rehabilitation by Elsevier, Netherlands.
-
2024 study published in Archives of Physical Medicine and Rehabilitation by Elsevier, Netherlands.
-
2025 study published in Progress in Rehabilitation Medicine by the Japanese Association of Rehabilitation Medicine, Japan.
-
-
2002 study published in Clinical Rehabilitation by SAGE Publications, United States.
-
2014 study published in Journal of Physiotherapy by Elsevier, Netherlands.
-
2013 Study Published by Stroke Journal, American Heart Association, United States
-
2006 study published in the Australian Journal of Physiotherapy by Elsevier, Netherlands.
-
2020 Study Published by Frontiers in Neuroscience
-
2022 Study Published by Annals of Geriatric Medicine and Research
-
2023 Study Published in the International Journal of Physiotherapy
-
2024 Study Published by The Rehabilitation Journal
-
2022 Study Published in International Journal of Environmental Research and Public Health,
-
2021 Study Published by Journal of International Medical Research
-
2025 study published by the Japanese Association of Rehabilitation Medicine.
-
2025 Study published by Cochrane Database of Systematic Reviews
-
2014 Study Published in Neurorehabilitation and Neural Repair.
-
Study Conducted on 93 Stroke People Published on Neurorehabilitation and Neural Repair
-
Study conducted on 197 Stroke people , published in Stroke Journal, American Heart Association, United States
-
2016 Study Conducted on 815 Stroke Survivors , Published by Journal of Rehabilitation Medicine
-
2025 Review Conducted on 933 People, Published by Disability and Rehabilitation, Taylor & Francis
-
2017 Study, Published in Systematic Reviews, BioMed Central
-
University of Glasgow Study Published in the American Heart Association journal Stroke
-
Study Conducted on 120 People After Stroke, Published in American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation
-
2023 Study Published in Pain and Therapy (Adis / Springer Nature)
-
2020 Study Published in 𝑭𝒓𝒐𝒏𝒕𝒊𝒆𝒓𝒔 𝒊𝒏 𝑵𝒆𝒖𝒓𝒐𝒔𝒄𝒊𝒆𝒏𝒄𝒆.
-
2016 Study Published in The Society of Physical Therapy Science, published by IPEC Inc., Japan
Disclaimer
This article is published for general information and educational purposes only. It does not constitute medical advice, health advice, diagnosis, or treatment of any kind.
All content on this website, including any studies, research, or information mentioned, is provided for informational purposes only and does not imply or guarantee any specific health outcomes.
We are not affiliated with, endorsed by, or connected in any way to any researchers, universities, research centres, journals, or institutions referenced in any article. No reference to any study, researcher, or publication should be interpreted as an endorsement or recommendation of any products sold or featured on this website.
Our products (and any products featured or linked in this article) are not intended to diagnose, treat, cure, prevent, or manage any disease or medical condition. Individual results may vary significantly. No guarantees or warranties are made regarding effectiveness, safety, or specific outcomes.
Always consult a qualified healthcare professional before using any product mentioned on this website or making any changes to your health routine, especially if you have a pre-existing medical condition, are pregnant, breastfeeding, have a pacemaker, or are taking medication.
To the fullest extent permitted by law, ORIEMS FIT, its directors, employees, authors, affiliates, and agents disclaim all liability for any loss, damage, cost, expense, or injury (whether direct, indirect, consequential, special, or incidental) arising from the use of, or reliance upon, any information in this article or from the purchase or use of any products featured or sold on this website.
© [ORIEMS FIT] – All Rights Reserved.
