NMES for Post-COVID ICU Muscle Loss: What the Evidence Shows

Neuromuscular electrical stimulation (NMES) can make a muscle contract when the person attached to it cannot move, cooperate, or even wake up. That makes it unusual in intensive care: a deeply sedated patient on a ventilator cannot lift a weight, but electrodes on the skin can still recruit muscle. Whether that preserves meaningful function after COVID-19 critical illness remains unsettled.

The clinical target is ICU-acquired weakness (ICUAW), a common consequence of severe illness, prolonged bed rest, inflammation, sedation, mechanical ventilation, and some ICU medications. About 46% of critically ill patients develop clinically significant weakness, with higher risk among people exposed to corticosteroids, sedatives, neuromuscular blockers, or mechanical ventilation for at least seven days. Muscle injury and atrophy can begin within 24 to 36 hours of sepsis onset. A review of mechanically ventilated adults reported measurable atrophy in roughly 37% of patients by ICU day 4.

Those numbers explain the appeal of NMES. They do not prove that it prevents post-intensive care syndrome (PICS), the long tail of physical, cognitive, and psychological problems that can follow critical illness.

What electrical stimulation can do in an ICU bed

NMES sends electrical current through surface electrodes placed over a target muscle. The current activates peripheral motor nerves and produces contractions without voluntary effort. In principle, that gives clinicians a way to load muscle during the period when voluntary rehabilitation is impossible.

Critical illness myopathy makes this logic especially compelling. It causes widespread wasting and paralysis of limb and trunk muscles, can delay liberation from mechanical ventilation, and can extend both ICU treatment and rehabilitation. The mechanism is plausible; the clinical question is whether enough externally produced contractions change outcomes that matter to patients.

Feasibility is relatively well established. In a randomized ICU trial focused on pressure-injury prevention, Baron et al. (2022) reported no serious adverse events attributed to NMES. That is reassuring, but deeply sedated patients cannot report pain or discomfort. Their safety monitoring must rely on observable and physiological signs, and that evidence remains less developed than safety data in awake rehabilitation patients.

COVID-19 evidence: a promising signal, not a settled treatment

The most directly relevant trial is Miyagishima et al. (2024), a controlled study of people with COVID-19 acute respiratory distress syndrome who received NMES under deep sedation. The trial found a lower incidence of ICUAW with NMES. Its setting matters: these were patients unable to create voluntary contractions, which is the clearest use case for the technology.

Still, the publication does not provide the magnitude of the ICUAW reduction in the available abstract, so readers cannot judge the absolute benefit, precision, or clinical size of the effect from that report alone. It is also the only COVID-specific controlled trial in this evidence base and appeared in Cureus, a journal whose publication model and peer-review standards warrant caution. One positive study is a reason to test the intervention better, not a reason to declare the question closed.

A 2025 systematic review by Carvajal-Tello et al. found only four post-COVID NMES studies. The studies assessed several outcomes, including ultrasound or cross-sectional measures of muscle mass, the Medical Research Council scale, PFIT-s, DEMMI, and SF-36 quality-of-life scores. But their protocols varied widely in electrode placement, frequency, pulse width, current intensity, session length, and number of treatment days. Those are not cosmetic differences; they may alter the contraction a patient receives. The review could not support a reliable, standardized post-COVID protocol or a firm pooled clinical conclusion.

Why the highest-level evidence remains inconclusive

The best available synthesis does not confirm that NMES prevents ICUAW. In an updated systematic review and meta-analysis of randomized trials, Nakanishi et al. (2023) concluded that it remains unclear whether NMES prevents ICU-acquired weakness. That conclusion should carry more weight than a collection of encouraging but heterogeneous signals.

Studies differ in more than their stimulation settings. Some test NMES alone, while others embed it in broader early-mobilization programs. Some measure muscle quality or strength scales; others measure ICU duration, ventilation duration, or functional recovery. A positive result on one endpoint does not establish a benefit on all the others.

For example, a 2024 network meta-analysis by Xu et al. found that early physiotherapy programs incorporating NMES were associated with improved muscle quality and shorter ICU stays and ventilation duration in mechanically ventilated adults. But the available findings did not isolate pooled NMES-specific estimates for muscle thickness or MRC-Sum Score. The benefit may reflect NMES, physiotherapy more broadly, or the package of care. It cannot be cleanly assigned to the electrodes.

The same attribution problem affects the ExPrES study, which combined NMES with exercise and high-protein supplementation in survivors of critical illness. Any improvement in muscle preservation or function could come from one component, their combination, or differences in participation. It does not establish an independent NMES effect.

Protocol may matter, but the dose is not known

The most useful dosing clue comes from Guerra-Vega et al. (2025), a randomized clinical trial in critically ill patients that compared medium-frequency with low-frequency NMES. Medium-frequency stimulation produced greater improvement in functional capacity during recovery. This is a real head-to-head signal that frequency matters.

It does not identify an optimal protocol. The available report does not provide the frequency values, pulse widths, session parameters, or effect estimates needed to prescribe a dose. Nor does the study settle the best time to start treatment relative to the first day of sepsis, when atrophy may already be underway. Clinical teams should not treat "NMES" as one uniform intervention when the delivered stimulus can vary substantially.

What adjacent evidence adds—and cannot add

Outside the ICU, a split-body randomized trial by Hardy et al. (2022) found that postoperative electrical muscle stimulation attenuated loss of muscle mass and function in the stimulated limb of older adults after major abdominal surgery, compared with the unstimulated limb of the same participant. That design controls for many differences between patients and supports the biological premise that electrically evoked contractions can blunt disuse-related decline.

It is still adjacent evidence. Major abdominal surgery and ICU ventilation are not interchangeable physiological states. Likewise, a review of interventions for people with motor-complete spinal cord injury found electrical stimulation among approaches used to build muscle in chronically paralyzed limbs, but the underlying studies carried a high risk of bias and involved a very different, chronic condition. These findings support plausibility, not a direct estimate of benefit for post-COVID ICU survivors.

What this means for patients and self-experimenters

NMES has a legitimate clinical rationale when illness makes ordinary exercise impossible. It may fit within a multimodal ICU or post-ICU rehabilitation plan, particularly as clinicians learn which stimulation parameters produce useful contractions and functional gains. It is not proven to prevent PICS, restore lost muscle on its own, shorten hospitalization, speed independent walking, or improve six-month function. Trials have not established those long-term outcomes for NMES as an isolated intervention.

That distinction matters outside the ICU too. A consumer muscle stimulator is not a substitute for resistance training, walking, nutrition, physical therapy, or medical evaluation after severe illness. Its most defensible role is narrower: assisted muscle activation during genuine forced immobility or rehabilitation, under clinical guidance when illness, surgery, nerve injury, skin problems, implanted devices, or cardiovascular concerns complicate the picture.

NMES can contract muscle when voluntary movement is impossible. The current evidence supports that rationale and suggests possible benefit, but the strongest ICU evidence still cannot confirm that it prevents ICU-acquired weakness.