Need help? Call us on
+44 (0)141 628 8222The first Cochrane review of exercise for fitness after spinal cord injury found that exercise may improve fitness, with low certainty. It says almost nothing about FES cycling, and the reason is the question it asked. Here is where FES earns its place, and where arm exercise does the job as well for less.
When we started to work with FES Cycling systems almost 20 years ago we were often asked whether FES cycling counts as the exercise a person with a spinal cord injury needs, and whether we could justify funding it on those grounds. In fact, there was quite supportive evidence of benefit and many home-based clients have adopted these systems over the years. As is typical in rehabilitation, however, there is always a need for further research and when you start to think about it, it's not always so easy to define exactly what exercise these individuals "need" and can access in practice. Physical exercise and training can have many benefits for both mind and body, but individuals with a spinal cord injury have differing needs and aspirations for exercise and fitness. Sometimes the best type of exercise you can do is the one that you can access and find the motivation to adhere to. This is not always so simple to define.
At the end of August, Cochrane published its first completed review on exercise for fitness after spinal cord injury. I expect it to be quoted in funding letters, in team meetings and in conversations with families. It deserves a careful reading first, because it answers a narrower question than most people will assume. On FES cycling in particular, it says very little, and the reason is not because it has no exercise value.
I expect most readers will know what a Cochrane review is, but just in case that's not you, here's a brief summary.
A Cochrane review is a systematic review of previously published healthcare research. This is useful because individual clinical trials can be small, use different outcomes, or produce conflicting results and a well-conducted review looks across these studies and can give a clearer and more balanced assessment of the state of the art. It's particularly useful in judging whether an intervention is likely to improve an outcome, for confirming the certainty of the evidence and for identifying gaps in knowledge. Because it's looking across a number of valid studies, it represents influential evidence, but it's not automatically the final answer to anything. Its usefulness depends on the quality and relevance of the available underlying studies. Unfortunately, rehabilitation often suffers from small and poor-quality studies.
I should declare an interest before going further. We supply FES cycling systems, including those based on the Pajunk Stim2Go, so you should read what follows knowing that. It's also why I want to be precise about what this review does and does not say, including the parts that don't help our case.
The review, by Ilha and colleagues with Lisa Harvey as senior author, asked whether physical exercise training improves cardiorespiratory fitness in people with spinal cord injury, compared with no intervention. The authors searched to December 2025 and included 17 randomised trials with 505 participants, although only 15 trials (337 people) contributed usable data. The training covered handcycling, rowing, wheelchair propulsion, robotic gait training, walking and running, and mixed gym equipment, two to five times a week, for anywhere from six weeks to nine months.
The headline result is that exercise training may increase cardiorespiratory fitness, with a standardised mean difference of 0.48 (95% confidence interval 0.18 to 0.77). In plain terms, that's a small to moderate effect. The certainty of the evidence was rated low, largely because 96% of the trials were at high risk of bias for the fitness outcome. Only one trial measured whether the effect lasted beyond the end of training.
Everything else was rated very low certainty: functional fitness, quality of life, depression and anxiety, and adverse events. The authors say plainly that their estimates for these outcomes "neither rule in nor rule out a therapeutic effect".
Strength was not an outcome of the review at all. Adverse events were poorly reported across the trials, and the only usable estimate came from a single trial and the events recorded there included autonomic dysreflexia, blood pressure changes, nausea, arthritis and pressure injuries. None of that is a reason not to exercise, but it is a reminder that the risks of training people with spinal cord injury have been counted carelessly.
Low certainty does not mean exercise doesn't work. It means the true effect could differ substantially from the estimate, mostly because of how the trials were run.
A larger 2023 meta-analysis by Hodgkiss and colleagues, which pooled 120 studies including 29 randomised trials, points the same way with a little more confidence. Exercise raised relative peak oxygen uptake by about 2.9 mL/kg/min over control, rated moderate certainty. Two reviews, built differently, agree on the direction.
For a funding letter, the practical reading is that the review supports exercise after spinal cord injury. It doesn't support any particular piece of equipment, including ours.
KEY POINT: Exercise probably improves fitness after spinal cord injury, with low certainty. On every other outcome the review can tell you very little, and it says so itself. It is a review of exercise, not of any device.
You will notice that FES cycling isn't in that list of training types. It would be easy to read this as Cochrane leaving FES out because there is no evidence of benefit, or finding against it. Neither is correct.
The review's protocol, published in 2024, explicitly allowed exercise "with or without functional electrical stimulation", including leg cycling and rowing. What it did not allow was a trial comparing one form of exercise with another. The control group had to receive no intervention, a placebo that could not plausibly affect fitness (stretching or skincare education, for example), or the same usual care as the exercise group.
That single rule filters out most of the FES cycling literature we have cited over the years.
The randomised FES trials that do exist mostly compare FES cycling with arm exercise, with passive cycling, or with a different FES programme. Those are sensible clinical questions, but they are not the question this review asked. Of the 30 references the authors excluded after reading the full text, seven went for having an ineligible control group.
One FES trial did get in, and what happened to it is instructive. Solinsky and colleagues, from Taylor's group in Boston, randomised people to hybrid FES rowing (arms and stimulated legs together), arms-only rowing, or a waiting list. The review combined the two rowing groups into a single exercise group and compared it with the waiting list, which is a reasonable way to handle a three-arm trial. It then classed the trial as upper-limb exercise. The result (a standardised mean difference of 0.50, with a confidence interval running from -0.21 to 1.22) therefore tells you about rowing in general, not about the stimulated legs. No leg-cycling FES trial was included, and FES isn't discussed anywhere in the review's findings or conclusions.
So the review's silence on FES cycling comes from the question it asked, not from a verdict on FES. The accurate way to put it to a colleague is that FES was eligible, but few FES trials are designed against a no-exercise control. The one that was included was pooled with arm exercise, so this review can't tell us anything about FES either way.
KEY POINT: The review is silent on FES cycling because it only accepted trials of exercise against no exercise, and FES trials are rarely designed that way. Silence by design is not a negative finding, and it isn't a positive one either.
This isn't the first time FES has fallen between two stools. The international exercise guidelines for adults with spinal cord injury, published by Martin Ginis and colleagues in 2018, recommend at least 20 minutes of moderate to vigorous aerobic exercise twice a week, plus strength training, for fitness, and at least 30 minutes of aerobic exercise three times a week for cardiometabolic health. They could not include FES. The panel stated that recommendations "could not be drafted" for exercise types with fewer than two high-quality trials, and named FES as one of them.
The 2021 systematic review by van der Scheer and colleagues was written partly to feed a first FES cycling guideline. It covered 92 studies and 999 adults and found improvements in power output and aerobic fitness in nearly all of the studies that measured them, but none of those studies was of the highest design levels, so the evidence for aerobic fitness was graded low. As far as I can find, that FES cycling guideline has not appeared in the five years since.
In other words, for eight years FES cycling has sat outside both the exercise guidelines and, now, the Cochrane review, for the same underlying reason which is simply not enough trials of the right design.
If the Cochrane review can't answer where FES fits, the trials that compare FES with other exercise can.
Hybrid cycling against handcycling. Bakkum and colleagues randomised 20 inactive people with long-standing spinal cord injuries to 16 weeks of either hybrid cycling (voluntary arm cranking plus FES leg cycling) or handcycling, twice a week for 30 minutes. Fitness changes were similar in both groups, and so were the changes in cardiometabolic risk factors, reported in a companion paper from the same trial. The authors concluded that the FES leg exercise added nothing they could measure over handcycling alone.
Arm crank against FES leg cycling. In a 16-week trial of 13 people training five days a week, Farkas, Gorgey and colleagues found that arm crank exercise produced greater gains in relative peak oxygen uptake, peak power and energy expenditure than FES leg cycling.
The broader picture. The Hodgkiss meta-analysis concluded that upper-body aerobic and resistance training appear the most effective options for fitness, and that cheaper modalities may produce similar or greater gains in relative peak oxygen uptake than FES.
Fitness is not the same as health. In a six-month trial of hybrid FES rowing, Solinsky and colleagues found that aerobic capacity rose but the prevalence of cardiometabolic disease did not change. The average dose actually achieved was 1.7 sessions a week, and only one of 40 participants met the exercise guidelines for the full six months.
Put plainly, for a person with good arm function whose goal is peak aerobic fitness, arm exercise can be at least as effective as FES leg cycling.
I read those trials with two eyes, though. Bakkum's participants trained at 65% to 75% of their heart rate reserve in both groups. If you prescribe intensity by heart rate, you are asking both groups' hearts to do the same work, so a similar change in fitness is roughly what you'd expect. That's my reading rather than the authors'. It suggests that the question hybrid exercise answers is not "does it train the heart better?" but "can this person reach the target at all?" For many people with good arms and shoulder function, the answer is yes without FES. For some, it is definitely not.
KEY POINT: For someone with good arm function whose goal is aerobic fitness, arm exercise matches or beats FES leg cycling assuming both are accessible. The case for FES lies elsewhere.
So where does it earn its place? In my reading, in four situations, and they are different from the one most marketing implies.
When the arms can't drive the heart. In tetraplegia and high thoracic injuries, there is often too little working muscle above the lesion to raise heart rate and oxygen uptake to a training level, however motivated the person. The Cochrane authors share the concern. They note that people with tetraplegia often show little improvement in fitness because they cannot generate enough workload, and they compared upper-limb with lower-limb and combined-limb exercise precisely because of doubts about whether arm exercise alone can recruit enough muscle. They found no difference between the two, but only four small studies (89 people) used the legs, none of them with FES, and the authors call the comparison not robust. The question is open, not answered.
This is where the physiology of FES leg cycling matters. In a classic study of eight people with tetraplegia, Hooker and colleagues matched oxygen uptake between arm cranking and FES leg cycling. Stroke volume was substantially higher during FES leg cycling than during arm cranking, which fits the idea that working leg muscles help return blood to the heart. Combining the two raised oxygen uptake by 54% over either alone. That was a single session, below peak effort, in eight people. It tells you about the stimulus, not about the training effect over months, and I'd present it that way. But the mechanism is sound, and it's why FES or hybrid exercise may be the only practical route to the guideline dose for some people with high-level injuries. At these levels, autonomic dysreflexia needs to be managed properly, and I've covered that in Autonomic Dysreflexia and FES Cycling: What Higher-Level Tetraplegics Need to Know Before Starting.
When you want the most from a session. A 2023 meta-analysis by Máté and colleagues, covering spinal cord injury and other central nervous system conditions, found that hybrid exercise produces higher peak oxygen uptake within a session than arm cranking or FES cycling alone. The longer-term studies were mostly uncontrolled, and Bakkum's trial found no added training benefit, so I would not claim hybrid exercise builds more fitness over months. It does give you more room to work with.
When the legs are the point. Muscle health is a different reason to use FES, and a legitimate one. In the van der Scheer review, muscle health was the only outcome graded high certainty. Arm exercise, however effective for the heart, does nothing for paralysed leg muscle. There is a caveat: the SCIPA Switch-On trial found no difference between FES cycling and passive cycling in muscle size when started within weeks of injury, so the timing and dose matter. I have covered what the muscle research measures in Does FES Cycling Build Muscle or Just Stop the Wasting? What the Research Measures.
When volume is the problem. The Solinsky dose figures are a reminder that in real life, people rarely reach the guideline dose, whatever the type of exercise. An exercise someone can do at home, seated and without transferring, may be the one that actually happens three times a week. That's an argument about adherence rather than physiology, and I've made the broader case in Volume vs Intensity in Neurological Rehabilitation: Why Consistency Often Matters More. In my experience, the most important consideration is to find a form of exercise that the individual can adhere to. Without that, every other discussion is just academic.
KEY POINT: FES cycling earns its place when the arms can't do the job alone, or when the legs themselves are the goal. Those are strong reasons, but they aren't the same as "FES is the best way to get fit".
If I were deciding where FES fits for an individual, I would start with four questions.
Can this person reach the guideline dose with their arms? If they have good arm function, shoulder integrity and fitness is the goal, arm ergometry, handcycling or wheelchair-based training comes first. NICE's guideline on rehabilitation after traumatic injury (NG211) says the same: consider upper-body aerobic training or seated exercise for people with limited lower limb mobility.
If not, what is limiting them? For someone with tetraplegia, or whose heart rate won't climb with arm work, FES or hybrid exercise may be the only way to get there. That is where I would put it first.
Are the legs themselves a goal? Leg muscle size and quality, and for some people spasticity management, are separate and excellent reasons to consider FES, and they should be argued on their own evidence, not on fitness.
Personal preference. As I've said a number of times in this article, sometimes the best exercise is the one a person will adhere to. With the arrival of Stim2Go, we found that not only do we have available an FES cycling system, but we have the means to augment many different forms of exercise with FES. This opens up the possibilities to think not just of FES cycling, but of a complete training and fitness regime.
A note on NICE, because I'm asked about it sometimes. NG211 mentions functional electrical stimulation, but as something to consider for mobility, upper limb function and walking, not for fitness. NICE has also published a Medtech Innovation Briefing on one FES cycling system, but a briefing is not a recommendation. Neither document recommends FES cycling for fitness, and I cannot cite either as if it did.
The next update of this review could say something about FES if trials existed to include. That needs FES trials against a usual-care or no-exercise control, large enough and well enough run to avoid the bias that sank the certainty of this one, and reporting the FES arm separately rather than pooling it with arm exercise. The authors' own research recommendations point the same way. We need fewer, larger, better-run studies, and deliberate inclusion of people with "very extensive weakness and paralysis", who are exactly the people for whom the case for FES is strongest and the alternatives are weakest.
That evidence probably isn't coming soon. Six trials are listed as ongoing, four of them in the UK, and judging by their titles none involves FES; two are explicitly arm-cycling or upper-body studies. If an FES trial of the right design appears, I'll update this article.
A first FES cycling guideline would help too. Until one exists, clinicians have to assemble the case themselves from a scattered literature, and case managers have to judge that case without a firm reference point. I hope articles like this one help.
If you have a patient or client for whom FES cycling is being considered for fitness, three things are worth doing with the clinical team.
Name the goal. Is it aerobic fitness, leg muscle health and appearance, or something else? The evidence for each is different, and so is the right equipment.
Test the arm route first in anyone with good upper limb and shoulder function, and measure what it achieves. If they reach the target dose, FES has to be justified on other grounds.
Measure what matters. Heart rate response, a simple exercise test, or sessions completed per week will tell you more about an individual than any pooled effect size. What is the individual going to be motivated to stick at and access without too much difficulty?
Where we come in is assessment. We look at the person's injury level, arm function and goals, try FES or hybrid exercise where it's practical, and give a realistic view of what it is likely to add. Sometimes the answer is that FES cycling will do something for this person that nothing else can. Sometimes it's that a handcycle should come first. Either is a useful answer. Modern products such as Stim2Go mean that we have many more opportunities to be creative with exercise and tailor programmes to suit the needs of each individual.
Decisions about any individual patient belong with the clinical team and the clinicians who assess them; nothing here replaces that. If you would like to talk a case through, I'm glad to, and a conversation costs nothing. I would rather help you ask the right question than see money spent answering the wrong one. When we started working with FES cycling almost 20 years ago, one of the biggest fears I had was that someone would purchase the equipment and not use it.
https://www.anatomicalconcepts.com
https://denervatedmuscle.com
Articles on this site are researched and drafted with the help of AI tools, then checked, edited and approved by me. Every citation is verified against the original source before publication.
Have questions about this article?
Get in TouchContinue reading about similar topics
Ask in your own words about FES cycling. Answers are drawn from both the patient guide and the clinical evidence review.
AI-generated answers based on the book. Not a substitute for professional medical advice. Conversations are stored for up to 90 days — please don't include personal details. Privacy policy
