What a Nerve Conduction Study Actually Tells You, and Why It Matters Before Electrical Stimulation

If you have a nerve injury, whether at the spine or further out in a limb, at some point, someone may send you for a nerve conduction study. You will lie on a couch while a neurophysiologist delivers small electrical pulses to your skin and, a little later, puts a fine needle into one or two muscles. A few weeks after that, a report lands with your consultant. It is dense, full of numbers and abbreviations, and it is written for one clinician to read to another. Almost nobody explains it to the person whose nerve it describes. Just this morning, I read a note from a client who said, "I had a nerve conduction test, which indicated severe injury to the nerves." You might wonder, like me, what severe means, and what do I do about it?

If a nerve conduction test isn't explained, it is a shame, because this single test answers the questions that determine what can be done next. In our work, what happens next might be some form of electrical stimulation, and the report tells us, more clearly than anything else available, whether stimulation can help you and which kind you would need.

I want to walk through what these studies measure, how they are read, and why the findings matter so much for the decision that follows. I am not a neurophysiologist and I do not perform these tests. We supply and advise on stimulation equipment, and I read a great many of these reports. My aim here is to give you enough understanding to follow the conversation your clinicians are having, and to ask better questions of your own.

Two tests, two different questions

People say "nerve conduction study" as a single thing, but a full examination is really two examinations that answer different questions.

The nerve conduction study proper looks at the "wiring". A nerve is stimulated at one point on the skin, and the response is recorded further along, either from the muscle that the nerve supplies or from the nerve itself. The size and speed of that response tell you how well the nerve is carrying signals.

The needle examination, or EMG, looks at the muscle. A fine needle records the electrical activity happening inside the muscle directly, both when the muscle is at rest and when you try to contract it. This is where the study finds the electrical fingerprints of a muscle that has lost its nerve supply.

Neither test on its own gives the full picture. Read together, they answer three questions that matter enormously for what comes next: is the nerve still physically connected, roughly how much of it is still working, and has the muscle been cut off from its nerve or merely cut off from the brain?

The nerve conduction study: how much of the wiring is left

The central measurement is deceptively simple. Stimulate a motor nerve, record from the muscle it supplies, and measure the size of the response. That response is called the compound muscle action potential, usually shortened to CMAP, and its amplitude is measured in millivolts.

Here is the useful part. The size of that response is roughly proportional to the number of nerve fibres still doing their job. A large response means a large working population of fibres. A small response means most of them have stopped conducting. An absent response means, to a first approximation, that the motor supply to that muscle is gone.

This is why comparing one side of the body with the other is the heart of the exercise. Suppose the nerve to a foot-lifting muscle on your affected side produces a response of 0.3 millivolts, while the same test on your unaffected side produces 3.1 millivolts. That comparison tells you, at a glance, that your affected side is running on roughly a tenth of its normal working fibres. That is a substantial loss. It is also, crucially, not zero. The distance between "a tenth" and "nothing" turns out to matter a great deal for what can be attempted.

KEY POINT: CMAP amplitude is not a literal headcount of nerve fibres, but it behaves like one. It provides a proportional measure of how much of the nerve is still working, and comparing the injured side with the healthy side is what gives that number meaning.

The study measures two other things worth knowing about. Latency and conduction velocity describe how fast the signal travels. A slow signal suggests damage to the insulating myelin sheath around the nerve. This kind of injury tends to recover well. A signal that is normal in speed but tiny in size points instead to a loss of fibres themselves, which is a more serious matter.

Then there are the sensory studies, which record from sensory nerves rather than motor ones, producing a small response called the SNAP. These matter more than you might expect, and I will come back to them, because they carry a clue about exactly where an injury sits.

The needle examination: what the muscle reveals

The needle examination is the part people can dread, and, in truth, it is uncomfortable rather than painful. It is also where some of the most valuable information comes from, because a muscle that has lost its nerve supply behaves in a very particular way, and the needle study can identify it.

At rest, a healthy muscle is electrically silent. A muscle that has recently lost its nerve supply is not. Individual muscle fibres, deprived of the nerve that used to command them, begin to twitch spontaneously and at random. These twitches are far too small to see or feel, but the needle picks them up as two characteristic signals: fibrillation potentials and positive sharp waves. Their presence is the clearest electrical evidence that nerve supply to that muscle has been lost. When you read that a muscle is "actively denervating", this spontaneous activity is what is meant.

On effort, the examiner watches how motor units recruit. A motor unit is a single nerve fibre together with all the many muscle fibres it controls, working as one team. A healthy muscle has hundreds of these teams, and when you push, more and more of them join in. After a nerve injury, there are fewer teams left to call on. When you try to contract the muscle as hard as you can, perhaps only a handful respond, or in a severe case, only one. The examiner calls this "reduced recruitment" or a "discrete" pattern, and it is a direct measure of how many working motor units survive.

The shape of each unit's signal tells a further story. When surviving nerve fibres begin to repair the damage, they grow small side branches and adopt muscle fibres that have been orphaned. A fibre that once commanded three hundred fibres might come to command far more. When that happens, the electrical signature of each motor unit grows larger, longer, and messier, described as "polyphasic". Seeing this is a direct sign that repair is underway.

KEY POINT: The needle examination distinguishes a nerve problem from a muscle problem, and it does more: it shows both the damage (spontaneous activity, fewer motor units) and the repair (enlarged, polyphasic units) in the same picture. It is the difference between a snapshot and a moving story.

Bruised, cut, or somewhere in between

Not all nerve injuries are equal, and the difference decides the outlook. For eighty years, clinicians have used an injury classification first set out by Herbert Seddon in 1943 and later refined by Sidney Sunderland. You do not need the details, but the three broad categories are genuinely useful for understanding.

Neurapraxia is the mildest. The nerve is bruised, and conduction is blocked at the site of injury, but the fibres themselves survive intact. This is the most recoverable kind, and recovery can be quick, often within weeks to a few months, because nothing has to regrow.

Axonotmesis is more serious. The nerve fibres themselves are damaged and their long tails degenerate, but the nerve's internal scaffolding survives. Fibres can regrow along their original path, but this only happens slowly (at roughly a millimetre a day), so recovery is measured in months and depends on how far the fibres have to travel.

Neurotmesis is the most severe. The nerve is effectively severed, scaffolding and all. Without surgical repair, meaningful recovery is not expected.

Real injuries are often a mixture, and Sunderland's five-grade system was designed to capture the subtleties in between the three descriptions above. A 2025 proposal from Susan Mackinnon's group has suggested extending it further, to a nought-to-six scheme organised around prognosis and recovery, a sign that this remains a living area rather than a settled history. For our purposes, the essential point is simpler: a nerve conduction study, especially when repeated, helps place an injury on this spectrum, and where it sits governs both the likely recovery and what is worth doing in the meantime.

Why timing changes everything

One of the most misunderstood things about these studies is that the answer depends heavily on when the test is done.

When a nerve fibre is cut off, its portion beyond the injury does not fail instantly. It degenerates over days, in a process known as Wallerian degeneration, after the physiologist who first described it. This has a direct consequence for the test. The spontaneous activity that marks denervation, those fibrillations and positive sharp waves, does not appear immediately. It takes roughly two to three weeks to develop, and the commonly quoted guideline is one to four weeks. It also appears earlier in muscles close to the injury and later in muscles further away, following the degeneration down the limb.

This is why a study done very early can understate an injury. A test in the first week or two can show what has been lost from the nerve conduction side, but it is too soon to show the muscle's denervation signals, and far too soon to show any signs of recovery, which take longer still to appear.

It is worth adding a note of professional candour here. That familiar "one to four weeks" figure rests largely on animal experiments from the middle of the last century rather than on modern human data, as a 2012 review in the journal Muscle and Nerve pointed out. The principle is sound and the timing is a reasonable working rule, but it is less precisely nailed down than usually suggested.

KEY POINT: A single early study captures loss, not recovery. This is the reason a repeat examination, often at around three months, is so informative. It is the point at which the first signs of repair, if any, become visible. Perhaps seen as larger and messier motor units, more of them firing, a growing response to nerve stimulation, and fewer fibrillations as orphaned fibres are reclaimed.

The distinction that matters most: has the muscle "lost its brain", or lost its nerve

For anyone weighing up electrical stimulation, this is the single most important thing a nerve conduction study establishes.

Every voluntary movement depends on a feature that acts like a two-stage relay. The upper motor neuron runs from the brain down through the spinal cord. The lower motor neuron leaves the cord and travels out along a peripheral nerve to the muscle. A weak or paralysed muscle can result from a break at either stage, and the two look similar from the outside but could not be more different underneath.

If the break is in the upper motor neuron, high in the relay, the muscle has lost its commander, but its own nerve structure is intact. On electrodiagnosis, this shows up as a preserved CMAP with no fibrillations and no positive sharp waves. The wiring from the spinal cord to the muscle still works; the muscle simply is not receiving orders from above. Most of the spinal cord injuries discussed publicly are of this type. There is paralysis but no denervation.

If the break is in the lower motor neuron, or in the nerve itself, the muscle has lost its connection to the world entirely. On electrodiagnosis, this presents as a reduced or absent CMAP, along with fibrillations and positive sharp waves. This is the signature of true denervation.

This maps onto the anatomy of a spinal injury in a way that can surprise people. Damage to the white matter of the cord, the long tracts running up and down, produces the upper motor neuron picture. Damage to the grey matter, where the lower motor neuron cell bodies reside, results in denervation.

Many real spinal injuries involve both, and there is emerging evidence, still from a small preliminary study and so to be treated as suggestive rather than settled, that lower motor neuron denervation in spinal cord injury can extend across several segments around the injury level, further than the textbook picture would predict. Injuries to the cauda equina or conus, at the lower end of the spine, produce a lower motor neuron picture almost by definition, because that is where the lower motor neurons live.

There is one more elegant piece of detective work worth knowing, because it shows how much a good report can deduce. Sensory nerve cell bodies sit in a small bundle just outside the spinal cord, called the dorsal root ganglion. If an injury is closer to the cord than that bundle, the sensory fibres farther out remain connected to their cell bodies and continue working, even though the person feels numb. So a report can identify normal sensory readings in a limb with no feeling. That apparent contradiction is not an error. It places the injury very close to the cord, on the motor side, and it is called a preganglionic pattern. It is often a more recoverable situation than injury further out, because the nerve is bruised or stretched rather than torn away.

KEY POINT: Upper motor neuron injury: the muscle has lost its commander, but its nerve still works. Lower motor neuron injury: the muscle has lost its nerve supply and is denervated. A nerve conduction study can confidently distinguish these, and they require completely different kinds of stimulation.

Why this decides the kind of stimulation

Everything above converges on one practical decision, and the physics of it is worth understanding.

Conventional stimulation, the kind used in most neuromuscular electrical stimulation (NMES) and functional electrical stimulation, including FES cycling, works through the nerve. It delivers short electrical pulses, typically a fraction of a millisecond (a few hundred microseconds) up to a millisecond long, repeated many times a second. Those short pulses are enough to trigger a working motor nerve, which then does the work of making the muscle contract. The arrangement is efficient precisely because the nerve sits ready and amplifies the signal for you. But it depends entirely on there being a working nerve to trigger. Where the nerve is gone, these external pulses have nothing to act on. Turning the machine up to maximum achieves nothing useful.

Denervated muscle needs a completely different approach. Without a nerve to amplify the signal, the pulse has to depolarise the muscle fibre membrane directly, and that requires roughly a thousand times more energy. It takes a much longer pulse: not a fraction of a millisecond, but tens to hundreds of milliseconds. Effective stimulation of denervated muscle uses long rectangular pulses of at least thirty milliseconds, or long triangular pulses of a hundred to five hundred milliseconds, at high currents that no consumer stimulator can produce. There is even a measurement, called chronaxie, that captures this: a chronaxie above one millisecond is itself a marker that a muscle has been denervated.

There is a hard consequence to this. A muscle that has been denervated for long enough gradually becomes inexcitable even to good conventional equipment, as its internal machinery falls into disorder. Within months, standard stimulation stops producing any contraction at all. This is not a fault in the equipment. It is the muscle telling you that it now needs the specialised, long-pulse approach.

KEY POINT: The nerve conduction study sorts you onto one of two paths. Intact nerve, weak muscle in which conventional stimulation can work through the nerve. Denervated muscle means only long-pulse, high-intensity, direct muscle stimulation has anything to work with. Choosing the wrong one wastes time that the muscle does not have.

A worked example

Let me make this concrete with a recent case, fully de-identified, because it shows the whole chain of reasoning in one picture.

A woman in her forties developed a drop foot immediately after routine lumbar spine surgery. She could not lift the front of her foot. Her nerve conduction study, done a couple of weeks later, found the following. The nerve to her main foot-lifting muscle produced a response of 0.3 millivolts on the affected side against 3.1 millivolts on the other: about a tenth of the normal signal, a large loss but not a total one. The needle examination found active denervation, fibrillations and positive sharp waves, and when she tried to lift her foot, only a single motor unit fired where there should have been hundreds. Her sensory readings, though, were normal, and she had no numbness. That normal sensory finding placed the injury very close to the spinal cord, on the motor side: a preganglionic pattern, affecting the nerve root, and a more recoverable kind of injury than it might first appear.

When I visited to try stimulation, the findings predicted exactly what happened. In conventional settings, a one-millisecond pulse at a typical frequency produced no contraction even at the highest current she could tolerate. With only a tenth of the nerve fibres conducting, there was too little for conventional stimulation to work with. When I switched to long pulses, fifty to two hundred and fifty milliseconds delivered slowly, one at a time, the muscle produced a visible twitch. The muscle was alive and responsive. It simply needed to be spoken to directly rather than through its depleted nerve.

The electrical study and the bedside demonstration agreed completely. That agreement is worth a great deal. It means the plan that follows is built on evidence rather than guesswork.

The window that makes this urgent

A muscle that has lost its nerve supply does not remain stable. It gradually loses bulk and tone, and over months and years, muscle tissue is replaced by fat and fibrous tissue, becoming simply incapable of contracting. If nerve fibres are going to regrow and reconnect, they need a muscle still healthy enough to reconnect to. The window for this is finite. Irreversible atrophy sets in over roughly twelve to eighteen months, after which reconnection, even if the nerve reaches the muscle, may no longer restore useful function. The decline is gradual rather than a cliff edge, but the direction is one way.

This is where electrical stimulation earns its place even when it cannot restore voluntary movement. Stimulation that makes a denervated muscle contract, even a twitch, keeps it in better condition through the waiting period. It preserves the muscle's bulk, its internal structure, and its blood supply, so that if reconnection happens there is something worth reconnecting to. And starting sooner is far easier than starting later, because there is less to reverse and more to protect.

The strongest evidence for this comes from the European RISE project and the Vienna group led by Helmut Kern, who spent two decades on the question of whether permanently denervated muscle could be rescued. In their home-based programme for people with complete lower motor neuron injuries, around ninety per cent of those who trained recovered or increased sustained contractions, and about a quarter regained enough force to stand during stimulation with support. That is not a return to walking, but it is a meaningful position recovered from complete flaccid paralysis, and it settles the question of whether the effort is worthwhile.

It is this study which inspired our own work with hundreds of individuals using the RISE stimulator and the KT series of products.

What a study can, and cannot, tell you

A nerve conduction study is powerful, and it has limits worth stating plainly.

It can tell you whether the nerve is still connected, roughly how much of it survives, whether the muscle is denervated, where the injury most likely sits, and, on a repeat test, whether recovery has begun. Those are exactly the facts needed to choose a stimulation strategy.

It cannot tell you, from a single early test, whether you will recover, because the signs of recovery take months to appear. It cannot see everything a scan can see: a study describes how the nerve is working, not what might be pressing on it, which is a question for an MRI. And the fibrillations that signal denervation, while highly reliable, are not unique to denervation, as one or two muscle diseases can produce them as well. None of this undermines the test. It simply means the study is one instrument in a set, most useful when its findings are read alongside the clinical picture and, where needed, imaging.

What I would suggest

If you or someone you advise has had a nerve injury and a nerve conduction study, a few practical points follow.

Ask your clinicians to explain the report in the terms described above. For example, is the nerve in continuity, roughly how much of it survives, is the muscle denervated, and where does the injury sit? Those answers shape everything.

Ask whether a repeat study, often at around three months, would be useful. It is frequently the single most informative thing available, because it shows whether recovery has started.

If the muscle is denervated, understand that time matters and that standard NMES stimulation equipment will not help. Denervated muscle needs specialised, long-pulse stimulation, and the sooner it starts, the more there is to preserve.

And if stimulation is appropriate, the choice between conventional and denervated-muscle equipment follows directly from the study. Getting that choice right is the difference between a programme that does something and one that quietly does nothing.

We work with both kinds of stimulation, and we read these reports as a matter of routine. If you have a study and are unsure what it means for your options, we are happy to talk it through in the context of your clinical team's advice. Decisions about your treatment belong with them. What we can do is help you understand what the findings mean for the equipment question and why.

Further reading

• Willmott AD, White C, Dukelow SP. Fibrillation potential onset in peripheral nerve injury. Muscle and Nerve 2012; 46(3): 332 to 340. https://doi.org/10.1002/mus.23310

• Pripotnev S, Mackinnon SE, et al. The Classification of Nerve Injury Revisited: Sunderland 0 to VI. 2025. https://pmc.ncbi.nlm.nih.gov/articles/PMC12550730/

• Kern H, Carraro U, et al. Home-based functional electrical stimulation for long-term denervated human muscle: history, basics, results and perspectives of the Vienna rehabilitation strategy. https://pmc.ncbi.nlm.nih.gov/articles/PMC4749003/

• Kern H, Carraro U, Adami N, et al. Home-based functional electrical stimulation rescues permanently denervated muscles in paraplegic patients with complete lower motor neuron lesion. Neurorehabilitation and Neural Repair 2010; 24(8): 709 to 721. https://doi.org/10.1177/1545968310366129

• Albeck MJ, et al. Segmental infralesional lower motor neuron abnormalities in patients with sub-acute traumatic spinal cord injury. medRxiv 2023. https://www.medrxiv.org/content/10.1101/2023.02.18.23286121

Related articles on our site

• Lower Motor Neuron Lesion Versus Upper Motor Lesion

• Normal Versus Denervated Muscle: Why the Rules of Electrical Stimulation Change After Nerve Injury

• Electrical Stimulation After Nerve Repair Surgery: When to Start and What to Expect

• Can I Start Electrical Stimulation Years After My Denervation Injury?

Our companion websites

fescycling.com
denervatedmuscle.com