An electromyography test, commonly called an EMG, is a diagnostic procedure that measures the electrical activity of your muscles and the nerves that control them. It helps doctors figure out whether symptoms like numbness, tingling, weakness, or pain originate from a nerve problem, a muscle disorder, or something else entirely. The test typically involves two parts that work together, and while it can be uncomfortable, the information it provides is difficult to get any other way.
Two Tests in One
When your doctor orders “an EMG,” what you’re actually getting is usually a two-part study. The full name for it is an electrodiagnostic study, and it combines nerve conduction studies with needle electromyography. Each part gives different information, and neither one alone paints the complete picture. Nerve conduction studies check how well electrical signals travel along your peripheral nerves, while the needle EMG portion looks at the electrical signals your muscles produce at rest and when you contract them. Together, they reveal whether a problem lies in the nerve, the muscle, or the junction between the two.
During the nerve conduction study, surface electrodes are placed on your skin over specific nerves and muscles. Small electrical pulses are sent through the electrodes to stimulate the nerve, and the response is recorded. The doctor can then measure how fast the signal travels and how strong it is. This part is especially useful for conditions where nerves are being compressed or damaged along their path.
The needle EMG portion works differently. A thin needle electrode is inserted directly into the muscle, where it picks up the electrical signals that muscle fibers produce. The doctor examines these signals while the muscle is relaxed, then asks you to gently contract the muscle and listens to the pattern of activity. Abnormal signals at rest, like spontaneous firing of muscle fibers, can indicate nerve damage. Unusual patterns during contraction can point to either nerve or muscle disease.
What Conditions It Helps Diagnose
EMG and nerve conduction studies are used to evaluate a broad range of neuromuscular problems. The conditions they’re most helpful for fall into a few major categories.
Carpal tunnel syndrome is one of the most common reasons people get sent for electrodiagnostic testing. Nerve conduction studies can confirm whether the median nerve is being compressed at the wrist, and how severely. Research has shown that the clinical severity of carpal tunnel syndrome correlates strongly with nerve conduction findings, with nighttime pain and numbness showing the strongest association with electrical abnormalities.1Archives of Hand and Microsurgery. Carpal tunnel syndrome: correlation of the severity of the clinical picture and electrophysiological studies Neurophysiological grading scales based on nerve conduction findings can classify carpal tunnel syndrome from very mild (detectable only with the most sensitive tests) all the way to extremely severe (where both sensory and motor nerve responses are essentially unrecordable).2Muscle & Nerve. A neurophysiological grading scale for carpal tunnel syndrome
Radiculopathy, the medical term for a pinched nerve root in the spine, is another frequent indication. If you have back pain shooting down your leg or neck pain radiating into your arm, EMG can help confirm whether a nerve root is involved and which one. The needle EMG portion is particularly important here. It has modest sensitivity, meaning it doesn’t catch every case, but high specificity, meaning that when it does find something, you can trust the result. It complements imaging like MRI, which can show structural problems but can’t tell you whether those problems are actually causing nerve dysfunction.3PubMed. Evaluation of persons with suspected lumbosacral and cervical radiculopathy: Electrodiagnostic assessment and implications for treatment and outcomes (Part II)
For serious conditions like amyotrophic lateral sclerosis (ALS), electrodiagnostic testing plays a critical role in reaching a diagnosis. The study helps exclude other treatable diseases that can mimic ALS and gathers evidence of widespread nerve damage across different body regions. Under current diagnostic criteria, the electrical findings from EMG are considered equivalent to clinical signs and symptoms when determining diagnostic certainty.4PubMed Central. Electrodiagnosis in persons with amyotrophic lateral sclerosis However, not every patient with ALS meets the classic electrical criteria on their first visit. In one study, roughly a third of ALS patients didn’t fulfill the standard electrodiagnostic criteria at initial presentation because their abnormalities weren’t distributed widely enough yet or their nerve conduction results were atypical.5PubMed. Role of electromyography in amyotrophic lateral sclerosis
Beyond these, EMG testing is used to evaluate peripheral neuropathies (generalized nerve damage, often from diabetes or other systemic conditions), myopathies (primary muscle diseases), neuromuscular junction disorders like myasthenia gravis, and various types of traumatic nerve injuries.
What Happens During the Test
The entire study usually takes between 30 minutes and an hour, depending on how many nerves and muscles need to be examined. You’ll typically sit or lie on an exam table, and the testing is performed by a neurologist, physiatrist, or specially trained technologist.
For the nerve conduction portion, the examiner attaches small adhesive electrodes to your skin and then uses a handheld stimulator to deliver brief electrical pulses over the nerve being tested. Each pulse feels like a quick, sharp tap or a jolt. Motor, sensory, and mixed nerves can each be tested by placing the recording electrodes in different positions: over a muscle supplied by the nerve, over a sensory nerve in the skin, or over the mixed nerve trunk itself.6Annals of Clinical Neurophysiology. Nerve conduction studies: basic principal and clinical usefulness The stimulation intensity is gradually increased until a full response is recorded, and several nerves may be tested on one or both sides of the body.
For the needle EMG, the examiner inserts a fine needle electrode into each muscle to be tested. You’ll be asked to relax completely while the doctor listens for abnormal spontaneous activity, then to contract the muscle gently while the pattern of motor unit firing is recorded. Each muscle is examined at multiple insertion sites during both rest and contraction.7Clinical Neurophysiology Practice. Needle EMG muscle identification: A systematic approach to needle EMG examination The number of muscles tested depends on what the doctor suspects. A straightforward carpal tunnel evaluation might involve only a few muscles, while a workup for ALS or a complex neuropathy could require sampling muscles across multiple limbs and even the back or tongue.
How Much Does It Hurt
This is the question most people actually care about, and the honest answer is that it’s uncomfortable but usually less painful than people expect. Research comparing patients’ anticipated pain to their actual experience found that patients consistently overestimated how much the needle EMG would hurt. The average expected pain score was higher than the average experienced pain score, and the nerve conduction portion was rated as less painful than the needle portion overall.8PubMed Central. Expected and Experienced Pain Levels in Electromyography
The nerve conduction study feels like repeated small electric shocks, which can be startling, but each pulse lasts only a fraction of a second. Some people find the stimulation over certain nerves (particularly the ulnar nerve at the elbow or the sural nerve near the ankle) more unpleasant than others. The needle EMG feels like a series of small pinches as the needle enters each muscle. Some muscles are more sensitive than others, and the discomfort can increase briefly if the needle hits a particularly tender area.
Discomfort during EMG is unavoidable, but it can be minimized. The single biggest factor affecting pain is which muscles need to be tested. Some interventions that may help include cooling sprays applied to the skin before needle insertion, over-the-counter anti-inflammatory medication like ibuprofen taken beforehand, and simple techniques the examiner can use such as quick insertion and stretching or tapping the skin before the needle goes in.9PubMed. Safety and pain in electrodiagnostic studies There is no need for general anesthesia, and local anesthetics are not typically used because they can interfere with the test results.
Risks and Complications
EMG is considered a very safe procedure. Serious complications are rare. A systematic review of published case reports and studies found 42 patients with reported complications across the entire medical literature, and none of them died. The most common complication was pneumothorax (a partially collapsed lung), which occurred in 17 of those patients, almost always when a needle was inserted into muscles near the chest wall or the base of the neck.10American Journal of Physical Medicine & Rehabilitation. Complications Associated With Electromyography
A large single-center study spanning 20 years documented complications during about 65,000 needle EMG examinations. The vast majority of problems were episodes of presyncope, the light-headed, woozy feeling that comes from a vasovagal response, the same reaction some people get when they see blood or feel pain. Actual fainting was rare, and a pneumothorax occurred once.11Open Access Journal of Neurology & Neurosurgery. Complications During Needle Electromyography; A Study Done for Twenty Years Bleeding disorders and skin reactions made up the remainder of reported complications in the broader literature, though many bleeding-related findings were subclinical, meaning they showed up on ultrasound but didn’t cause symptoms.
Blood Thinners, Pacemakers, and Other Concerns
One of the most common worries patients have is whether they need to stop blood-thinning medications before the test. The evidence here is reassuring. Studies using ultrasound to look for hematomas after needle EMG in patients taking warfarin (with elevated INR values) or antiplatelet medications like aspirin and clopidogrel found that hematoma formation was uncommon and, when it did occur, the hematomas were small and caused no symptoms. In one study, only two small subclinical hematomas were found among patients on warfarin, and one was found in a patient on clopidogrel, with none in the control group.12PubMed. Complications of needle electromyography: hematoma risk and correlation with anticoagulation and antiplatelet therapy Applying direct pressure to the insertion site after removing the needle appears to keep the risk low.13Nature Reviews Neurology. Is needle electromyography safe in patients on anticoagulation or antiplatelet therapy? Most labs do not require patients to stop these medications before testing, though you should always tell your doctor what you’re taking.
Patients with implanted cardiac devices like pacemakers and defibrillators also wonder whether the electrical stimulation from nerve conduction studies could interfere with their devices. Studies have specifically tested this and found that the electrical impulses used in standard nerve conduction studies were never detected by the sensing circuitry of pacemakers or implantable cardioverter-defibrillators. The programmed settings of the devices were unaffected.14PubMed. Safety of nerve conduction studies in patients with implanted cardiac devices 15PubMed. Safety of nerve conduction studies in patients with implantable cardioverter-defibrillators
Timing Matters
If you’ve had a nerve injury, getting tested too early can produce misleading results. When a nerve is cut or severely damaged, it takes time for the downstream effects to become visible on EMG. The muscle fibers that have lost their nerve supply don’t start showing the characteristic spontaneous electrical activity (fibrillation potentials) until roughly two to three weeks after the injury, and sometimes longer depending on how far the injury site is from the muscles being tested.
For traumatic nerve injuries where the question is whether the nerve will recover on its own or needs surgical repair, research on radial nerve injuries found that needle EMG achieves a specificity above 95% for determining lesion severity starting around four months after injury. If no voluntary motor unit activity is found at that point, it can be considered a reliable indicator that surgical exploration is warranted.16PubMed. Optimal timing of needle electromyography to diagnose lesion severity in traumatic radial nerve injury Testing too early can show denervation that might still recover, while testing at the right window gives much more actionable information.
For conditions like carpal tunnel syndrome or radiculopathy, timing is less critical because these tend to be chronic or subacute problems. But even there, a completely normal EMG doesn’t always mean nothing is wrong. The sensitivity of nerve conduction studies for carpal tunnel syndrome in one study was only about 49%, meaning roughly half of patients with clinical symptoms had normal test results.17PubMed Central. Suspected carpal tunnel syndrome: Do nerve conduction study results and symptoms match? This is why doctors treat the whole picture rather than relying solely on test numbers.
Testing in Children
EMG testing can be performed on infants, children, and adolescents, and it continues to be valuable for diagnosing childhood neuromuscular disorders. The diagnostic questions are often the same as in adults (is there a neuropathy, a myopathy, a neuromuscular junction problem?) but the practical challenges are different. Young children can’t always cooperate with instructions to relax or gently contract a muscle, and they may be frightened by the procedure. A consensus panel of specialists confirmed that electrodiagnostic studies have high utility in the pediatric population, while emphasizing the importance of using age-appropriate reference values, since nerve conduction speeds and other measurements change as children grow.18PubMed. Utility and practice of electrodiagnostic testing in the pediatric population: An AANEM consensus statement
In very young or uncooperative children, examiners sometimes adapt the protocol. The needle EMG portion can be performed during sleep if spontaneous activity at rest is the primary question. Some labs use sedation for prolonged studies, though this eliminates the ability to test voluntary muscle contraction. Parents should ask ahead of time what accommodations the lab makes for younger patients.
How to Prepare
Preparation for an EMG is straightforward, but there are a few things that can affect test quality. On the day of the test, don’t apply lotions, creams, or oils to your skin. These create a barrier between the electrodes and your skin that can degrade the signal and make the nerve conduction portion less accurate. If you normally use moisturizer, skip it on the areas likely to be tested (arms, legs, hands, feet).
Wear loose, comfortable clothing or bring shorts and a short-sleeved shirt, since the examiner will need access to your limbs. If your test involves the legs and back, a gown may be provided. Let the lab know in advance about any implanted cardiac devices, bleeding disorders, or medications you take, even though these are unlikely to prevent testing. If you have a strong history of fainting during medical procedures, mention it so the staff can take precautions like keeping you lying down throughout the test.
There’s no need to fast. You can eat, drink, and take most of your usual medications beforehand. Some medications used for nerve pain, like gabapentin or pregabalin, don’t need to be stopped. Pyridostigmine, used for myasthenia gravis, may occasionally be held before testing if the doctor is specifically evaluating neuromuscular junction function, but your prescribing physician will tell you if that applies to you.
Surface EMG Outside the Clinic
The type of EMG used in diagnostic medicine involves needle electrodes, but there is a different version that uses only surface electrodes stuck to the skin, called surface electromyography (sEMG). This non-invasive approach is not considered a replacement for needle EMG in diagnosing specific neuromuscular diseases. A review by the American Association of Neuromuscular and Electrodiagnostic Medicine found that while surface EMG may be useful for detecting the general presence of neuromuscular disease, there isn’t enough evidence to support using it to distinguish between nerve and muscle disorders or to diagnose specific conditions.19PubMed. American Association of Neuromuscular & Electrodiagnostic Medicine evidenced-based review: use of surface electromyography in the diagnosis and study of neuromuscular disorders
Where surface EMG does thrive is in research and applied settings outside the diagnostic lab. It has been widely adopted in sports science, ergonomics, and physical rehabilitation as a way to assess muscle fatigue during exercise and work tasks.20PubMed Central. Application of Surface Electromyography in Exercise Fatigue: A Review Because it’s painless and non-invasive, it lends itself to repeated measurements during physical activity. Researchers and ergonomists use it to study things like how quickly a muscle fatigues during a factory task, whether a particular lifting technique puts excessive strain on the lower back, or how an athlete’s muscle firing pattern changes as they tire.21PubMed. Surface EMG based muscle fatigue evaluation in biomechanics If you see consumer wearable devices advertising “EMG-based” muscle tracking during workouts, they’re using this surface approach, not the clinical needle version.
Artificial Intelligence and EMG Interpretation
Interpreting EMG signals has traditionally been a skill-dependent task, relying heavily on the experience of the examiner who listens to and visually analyzes the electrical patterns. Researchers have been exploring whether artificial intelligence could assist with or automate parts of this process. A scoping review identified 51 studies that used machine learning or deep learning to classify needle EMG signals, with most focusing on distinguishing between healthy signals and those from patients with ALS or myopathy. Many of these studies reported high accuracy, but the review concluded that current AI models are not sufficient for clinical use, in part because most were trained on a single small open-source dataset of only 25 subjects and were vulnerable to bias and overfitting.22PubMed. Artificial intelligence for automatic classification of needle EMG signals: A scoping review
Efforts to use larger, real-world clinical databases have shown more modest accuracy figures. One study using a large clinically acquired database reported about 62% accuracy for classifying EMG signals into clinical categories, a far cry from the 90%-plus numbers seen in studies trained on small curated datasets.23PubMed. Electromyography Signal Classification With Artificial Intelligence for Detection of Neuromuscular Disorders Using a Large Clinically-Acquired Database The gap between performance on tidy research data and messy clinical data is a familiar problem in medical AI, and it’s one reason EMG interpretation still belongs firmly in the hands of trained clinicians. The technology has promise as a potential decision-support tool down the road, but for now, the human ear and eye remain the standard.