NCS & EMG Interpretation
Electrodiagnostic (EDX) testing — comprising nerve conduction studies (NCS) and needle electromyography (EMG) — is considered an extension of the neurologic examination and represents the most important diagnostic tool for evaluating peripheral nerve and muscle disorders. NCS assess the integrity of peripheral nerves by recording electrical responses to controlled stimulation, while needle EMG evaluates the electrical activity of muscle fibers at rest and during voluntary contraction. Together, these studies can localize lesions along the motor unit, distinguish axonal from demyelinating pathology, differentiate neurogenic from myopathic processes, determine chronicity, and guide the clinician toward a specific diagnosis. Mastery of EDX interpretation is fundamental to neuromuscular practice.
Bottom Line
- NCS distinguish axonal from demyelinating injury: Reduced amplitudes indicate axonal loss, while prolonged distal latencies, slowed conduction velocities, and conduction block indicate demyelination
- Needle EMG provides temporal information: Fibrillation potentials appear 2–3 weeks after denervation; chronic reinnervation produces large, long-duration, polyphasic motor unit potentials (MUPs) over months
- Myopathic MUPs are small in amplitude, short in duration, and polyphasic, with early (rapid) recruitment at low force levels
- Conduction block (a ≥50% drop in proximal CMAP amplitude) is the hallmark of acquired demyelinating neuropathies and distinguishes them from hereditary forms
- Pattern recognition integrates NCS and EMG findings to differentiate polyneuropathies, motor neuron disease, radiculopathies, entrapment neuropathies, and myopathies
- Common pitfalls include cold limb temperature (falsely slows conduction velocity), performing EMG too early after injury (before fibrillations appear), and Martin-Gruber anastomosis (mimics conduction block)
Nerve Conduction Studies
NCS record the electrical responses generated by peripheral nerves when stimulated with brief electrical impulses. Motor NCS record compound muscle action potentials (CMAPs) from muscle, while sensory NCS record sensory nerve action potentials (SNAPs) directly from sensory nerve fibers. Each measurement provides distinct information about nerve integrity.
Motor NCS Parameters
- CMAP amplitude: Reflects the number of functioning motor axons innervating the recorded muscle; reduced amplitude indicates axonal loss (or distal conduction block)
- Distal motor latency (DML): The time from distal stimulation to CMAP onset; prolongation (>130% of the upper limit of normal) suggests distal demyelination
- Conduction velocity (CV): Calculated from stimulation at two sites along the nerve; slowing below 70–80% of the lower limit of normal indicates demyelination; mild slowing (70–80% of normal) can occur secondary to large-fiber axonal loss
- F-waves: Late responses generated by antidromic activation of motor neurons; prolonged F-wave latencies or absent F-waves suggest proximal demyelination (e.g., at nerve roots) and are important in early Guillain-Barré syndrome (GBS)
- H-reflex: The electrophysiologic equivalent of the ankle jerk reflex; assesses the S1 sensory-motor arc; absence or prolongation supports S1 radiculopathy or polyneuropathy
Sensory NCS Parameters
- SNAP amplitude: Reflects the number of functioning sensory axons; reduced amplitude indicates sensory axonal loss
- Sensory conduction velocity: Slowing suggests sensory nerve demyelination; often the earliest abnormality in entrapment neuropathies (e.g., carpal tunnel syndrome)
- Key principle: SNAPs are normal in preganglionic lesions (radiculopathy) because the dorsal root ganglion cell body and its distal axon remain intact; SNAP abnormalities therefore exclude a pure radiculopathy and indicate a lesion distal to the dorsal root ganglion
Axonal vs. Demyelinating Patterns
The distinction between axonal loss and demyelination is one of the most critical determinations in electrodiagnostic medicine. This distinction drives the differential diagnosis and guides subsequent management.
| NCS Feature | Axonal Loss | Demyelination |
|---|---|---|
| CMAP amplitude | Reduced (proportional to axonal loss) | Normal or mildly reduced (unless conduction block present) |
| SNAP amplitude | Reduced | Normal or mildly reduced |
| Distal motor latency | Normal or mildly prolonged | Prolonged (>130% upper limit of normal) |
| Conduction velocity | Normal or mildly slow (≥70–80% of lower limit) | Significantly slow (<70–80% of lower limit) |
| Conduction block | Absent | Present (acquired demyelination) |
| Temporal dispersion | Absent | Present (acquired demyelination) |
| F-wave latencies | Normal or absent (if axonal loss severe) | Prolonged or absent |
| Needle EMG | Fibrillations, positive sharp waves; large, long-duration MUPs | May be normal; fibrillations only if secondary axonal loss |
Conduction Block & Temporal Dispersion
Conduction block and temporal dispersion are the electrophysiologic signatures of acquired demyelinating neuropathies and have important diagnostic implications.
- Conduction block: Defined as a ≥50% reduction in proximal CMAP amplitude (or area) compared to the distal CMAP, with <30% change in duration. This indicates that action potentials fail to propagate through a segment of demyelinated nerve. Conduction block is the hallmark of acquired demyelinating neuropathies (CIDP, multifocal motor neuropathy, GBS) and is essentially absent in hereditary demyelinating neuropathies such as Charcot-Marie-Tooth type 1
- Temporal dispersion: An abnormal increase in CMAP duration (>30% prolongation of the proximal compared to distal response) caused by differential slowing of conduction across individual nerve fibers within the same nerve. The asynchronous arrival of action potentials at the recording electrode produces a wider, lower-amplitude, polyphasic waveform. Temporal dispersion is seen in both acquired and hereditary demyelinating neuropathies
- Distinguishing block from dispersion: In pure temporal dispersion, the CMAP area is relatively preserved despite amplitude reduction and duration prolongation. In conduction block, there is a true loss of CMAP area because some axons completely fail to conduct through the demyelinated segment
Sural Sparing Pattern
- In early GBS, sensory NCS may show absent or reduced median and ulnar SNAPs while the sural SNAP is preserved — the "sural sparing pattern"
- This occurs because demyelination preferentially affects longer, larger-diameter nerves in the upper extremities before affecting the shorter sural nerve
- The sural sparing pattern has a specificity of >90% for acute inflammatory demyelinating polyneuropathy (AIDP) when present in the appropriate clinical context
- It can be the earliest electrophysiologic abnormality in GBS, sometimes detectable within the first few days of weakness
- The reverse pattern (abnormal sural with normal upper extremity SNAPs) is characteristic of length-dependent axonal polyneuropathy
Needle Electromyography
Needle EMG evaluates the electrical activity of individual muscles by inserting a concentric or monopolar needle electrode into the muscle. The examination assesses three phases: insertional activity, spontaneous activity at rest, and voluntary motor unit potential (MUP) analysis with recruitment assessment.
Insertional Activity
- Normal: Brief burst of electrical activity lasting <300 ms that ceases when needle movement stops
- Increased insertional activity: Prolonged electrical discharges beyond needle movement; indicates membrane instability seen in acute denervation, inflammatory myopathies, and early reinnervation
- Decreased insertional activity: Reduced or absent response to needle insertion; indicates loss of viable muscle tissue from fibrosis, severe atrophy, or end-stage muscle replacement (e.g., fatty infiltration in chronic myopathy or long-standing denervation)
Spontaneous Activity
In normal muscle at rest, no electrical activity should be recorded outside the endplate region. The presence of spontaneous activity is abnormal and provides critical diagnostic information.
| Discharge Type | Morphology | Clinical Significance |
|---|---|---|
| Fibrillation potentials | Brief, biphasic or triphasic spikes; regular firing rate (1–30 Hz) | Denervation (appear 2–3 weeks post-injury); also in inflammatory myopathies, necrotizing myopathies, and NMJ disorders |
| Positive sharp waves (PSWs) | Initial sharp positive deflection followed by slow negative phase; regular firing | Same significance as fibrillation potentials; often appear together; represent spontaneous depolarization of single muscle fibers |
| Fasciculation potentials | Motor unit–sized potentials firing irregularly at low rates | Benign fasciculations are common in healthy individuals; pathologic when accompanied by fibrillations, MUP remodeling, or reduced recruitment (motor neuron disease, radiculopathy, chronic neuropathy) |
| Myotonic discharges | Repetitive, waxing-and-waning discharges with characteristic "dive-bomber" sound | Myotonic dystrophy, myotonia congenita, paramyotonia congenita, proximal myotonic myopathy (DM2), acid maltase deficiency, and some toxic myopathies (statins, colchicine) |
| Complex repetitive discharges (CRDs) | Uniform, polyphasic potentials firing at a fixed rate; abrupt onset/cessation ("machine-like") | Nonspecific; chronic denervation, chronic myopathies, radiculopathy, motor neuron disease; reflect ephaptic transmission between adjacent denervated muscle fibers |
| Myokymic discharges | Grouped, rhythmic bursting of motor unit potentials ("marching soldiers") | Radiation plexopathy (highly characteristic), GBS, MS, chronic nerve compression |
| Neuromyotonic discharges | Very high frequency (150–300 Hz) decrementing bursts | Peripheral nerve hyperexcitability (Isaac syndrome), Morvan syndrome, anti-CASPR2 antibodies |
Motor Unit Potential Analysis
The morphology and behavior of MUPs during voluntary activation provide essential information about whether the underlying process is neurogenic (neuropathic) or myopathic.
| MUP Feature | Neuropathic (Neurogenic) | Myopathic |
|---|---|---|
| Amplitude | Increased (due to collateral sprouting; more muscle fibers per motor unit) | Decreased (fewer viable muscle fibers per motor unit) |
| Duration | Increased (broader motor unit territory from reinnervation) | Decreased (smaller motor unit territory) |
| Phases | Polyphasic (especially during active reinnervation); >4 phases is abnormal | Polyphasic (due to temporal dispersion of muscle fiber potentials within the motor unit) |
| Recruitment | Reduced (decreased number of recruitable motor units; remaining units fire at high rates) | Early / rapid (many motor units recruited at low force to compensate for individual unit weakness) |
| Firing rate | Increased rate of individual units before additional units recruited (≥15–20 Hz) | Normal rate but excessive number of units activated for the force generated |
| Stability | Unstable (moment-to-moment variation) during active reinnervation; stable in chronic neurogenic changes | Unstable in inflammatory myopathies; variably stable in chronic myopathies |
EDX Pattern Recognition
Integrating NCS and needle EMG findings into recognizable patterns is the key skill in electrodiagnostic interpretation. The following table summarizes the characteristic EDX profiles of the most commonly encountered neuromuscular conditions.
| Condition | Motor NCS | Sensory NCS | Needle EMG |
|---|---|---|---|
| Axonal polyneuropathy | Low CMAPs distally; normal or mildly slow CV | Low SNAPs, length-dependent; sural often worst | Fibrillations/PSWs in distal muscles; large, long-duration MUPs; reduced recruitment distally |
| Demyelinating polyneuropathy (acquired — CIDP) | Slow CV; prolonged DML; conduction block; temporal dispersion; prolonged F-waves | Prolonged latencies; may be normal early | Normal or neurogenic changes in weak muscles; fibrillations if secondary axonal loss |
| Demyelinating polyneuropathy (hereditary — CMT1) | Uniformly slow CV (<38 m/s in upper extremities); no conduction block | Uniformly slow or absent SNAPs | Chronic neurogenic MUPs; no or minimal fibrillations |
| Motor neuron disease (ALS) | Low CMAPs (especially in atrophic muscles); normal CV; normal DML | Normal (critical feature) | Widespread fibrillations, PSWs, and fasciculations across ≥3 regions; large, long-duration, unstable MUPs; reduced recruitment |
| Radiculopathy | Usually normal; CMAP may be low if severe axonal loss in one myotome | Normal (preganglionic lesion) | Fibrillations/PSWs in myotomal distribution including paraspinals; neurogenic MUPs in affected myotome; normal in non-affected myotomes |
| Entrapment / mononeuropathy | Focal slowing or conduction block at entrapment site; may have low distal CMAP | Slow CV or low SNAP across the entrapment site | Neurogenic changes limited to muscles distal to the entrapment; paraspinals normal |
| Myopathy | Normal or mildly low CMAPs (in severe cases); normal CV | Normal | Short-duration, low-amplitude, polyphasic MUPs; early recruitment; fibrillations/PSWs in inflammatory or necrotizing myopathies |
| NMJ disorder (MG) | Normal or low CMAP; ≥10% decrement on 2–3 Hz repetitive nerve stimulation | Normal | Unstable (varying) MUPs; increased jitter on single-fiber EMG; no fibrillations |
Acquired vs. Hereditary Demyelination
- Acquired (CIDP, GBS, MMN): Non-uniform (segmental) slowing across different nerves and nerve segments; conduction block and temporal dispersion are common; slowing is asymmetric between nerves
- Hereditary (CMT1): Uniform slowing across all nerves and segments (typically <38 m/s in upper extremities); conduction block is absent; slowing is symmetric
- This distinction is critical because acquired demyelinating neuropathies are treatable with immunomodulatory therapy (IVIG, plasma exchange, corticosteroids), while hereditary demyelinating neuropathies are not
- In practice, the presence or absence of conduction block is the single most useful feature for distinguishing acquired from hereditary demyelination
Recruitment Analysis
Recruitment — the orderly activation of motor units with increasing voluntary effort — is one of the most informative components of the needle EMG examination. It follows Henneman's size principle, whereby smaller motor units are recruited first, followed by progressively larger units as force increases.
- Normal recruitment: At full effort, a complete interference pattern is seen with many overlapping MUPs obscuring the baseline
- Reduced (neurogenic) recruitment: Fewer motor units are available, so existing units fire at abnormally high rates (≥15–20 Hz) before additional units are recruited; the firing rate-to-number of units ratio is increased. At maximal effort, a discrete or reduced interference pattern is seen with individual MUPs still identifiable against the baseline
- Early (myopathic) recruitment: Individual motor units generate less force because of muscle fiber loss, so the CNS recruits an excessive number of units at low force levels; a full interference pattern appears with minimal effort. The pattern is sometimes described as "full but low amplitude"
- Poor activation: Reduced firing due to upper motor neuron lesion or poor effort (not a peripheral neuromuscular process); the firing rate remains low and additional units are not activated, but MUP morphology is typically normal
Timing of EDX Studies After Acute Injury
The timing of electrodiagnostic evaluation relative to the onset of a nerve injury or neuropathic process significantly affects the findings and must be considered in interpretation.
| Time After Injury | Expected NCS Findings | Expected EMG Findings |
|---|---|---|
| 0–3 days | CMAP and SNAP may still be normal distally (Wallerian degeneration not yet complete); conduction block may be demonstrable at the injury site | Reduced recruitment only (loss of voluntarily activated motor units); no fibrillation potentials |
| 3–7 days | Distal CMAP and SNAP begin to decline as Wallerian degeneration progresses (motor fibers degenerate by day 3–5; sensory by day 6–10) | Fibrillations may appear in paraspinal muscles (shorter distance from injury); still absent in limb muscles |
| 2–3 weeks | CMAP and SNAP amplitudes reach their nadir, reflecting the full extent of axonal loss | Fibrillation potentials and PSWs appear in denervated limb muscles; this is the optimal time for initial EDX evaluation in most acute injuries |
| Months | Amplitudes may improve with reinnervation | Chronic neurogenic MUP changes (increased amplitude, duration); fibrillations may persist or resolve with successful reinnervation |
Common Pitfalls in EDX Interpretation
- Cold limb temperature: Limb temperature <32°C (hand) or <33°C (leg) artificially slows conduction velocity by 1.5–2.5 m/s per degree Celsius, prolongs distal latencies, and increases SNAP and CMAP amplitudes (by slowing sodium channel inactivation, widening the action potential). Cold extremities can cause a normal nerve to meet "demyelinating" criteria, leading to misdiagnosis of CIDP. Always warm limbs before testing and monitor skin temperature throughout the study
- Performing EMG too early: Fibrillation potentials do not appear for 14–21 days after axonal injury in limb muscles (and 7–10 days in paraspinal muscles). An EMG performed within the first week after an acute nerve injury will miss denervation potentials and may be falsely interpreted as normal. The ideal timing for initial evaluation is ≥3 weeks post-injury
- Martin-Gruber anastomosis (MGA): An anomalous motor fiber crossover from the median nerve to the ulnar nerve in the forearm, present in 15–30% of the population. MGA produces a larger proximal than distal CMAP on median motor studies (or a smaller-than-expected distal CMAP) and can mimic a conduction block on ulnar motor studies (the ulnar CMAP at the elbow appears larger than at the wrist because the crossover fibers rejoin the ulnar nerve proximally). Failure to recognize MGA leads to erroneous diagnoses of median neuropathy, ulnar conduction block, or spurious conduction velocity calculations
- Axonal loss mimicking demyelination: Severe axonal loss can produce mild conduction velocity slowing (loss of the fastest-conducting large-diameter fibers) and may be misinterpreted as a demyelinating process. A general guideline: conduction velocity should not fall below 70–80% of the lower limit of normal from axonal loss alone. If it does, a primary demyelinating process should be suspected
- Phase cancellation: In sensory NCS, the biphasic waveform of individual nerve fiber action potentials leads to cancellation when they overlap out of phase. This physiologic phenomenon disproportionately reduces SNAP amplitude with increasing nerve length, making it difficult to compare proximal and distal sensory NCS directly
- Co-stimulation and volume conduction: Inadvertent stimulation of adjacent nerves (e.g., stimulating the ulnar nerve while recording from the median nerve) or volume-conducted potentials from nearby muscles can produce spurious responses. Careful electrode placement and stimulus adjustment are essential
- Edema and obesity: Subcutaneous tissue increases the distance between the stimulating/recording electrode and the nerve, reducing recorded amplitudes and potentially prolonging latencies. This can mimic axonal loss or focal neuropathy, particularly at common sites such as the carpal tunnel
Special Considerations
Late Responses: F-Waves and H-Reflexes
Late responses provide information about proximal nerve segments that are otherwise difficult to assess with standard motor and sensory NCS.
- F-waves are generated by antidromic activation of a small percentage of motor neurons, producing a low-amplitude, variable-morphology late response. Prolonged F-wave minimum latency, increased F-wave chronodispersion, or absent F-waves suggest proximal demyelination and are among the earliest abnormalities in GBS
- H-reflexes represent the electrophysiologic equivalent of the monosynaptic stretch reflex. The tibial H-reflex is routinely recorded from the soleus muscle and assesses the S1 arc. Absent or prolonged H-reflexes are seen in S1 radiculopathy and polyneuropathies; bilateral absence in older adults (>60 years) may be a normal variant
Repetitive Nerve Stimulation
While not part of standard NCS/EMG, repetitive nerve stimulation (RNS) is an essential extension when a neuromuscular junction disorder is suspected. Low-frequency (2–3 Hz) RNS demonstrating a ≥10% decrement in CMAP amplitude by the 4th or 5th stimulus is characteristic of postsynaptic NMJ disorders (myasthenia gravis). An incremental response (≥100% increase) at high-frequency stimulation (20–50 Hz) or after brief exercise is the hallmark of presynaptic NMJ disorders (Lambert-Eaton myasthenic syndrome).
Single-Fiber EMG
Single-fiber EMG (SFEMG) is the most sensitive test for NMJ dysfunction. It measures "jitter" — the variability in the time interval between two muscle fiber action potentials belonging to the same motor unit. Increased jitter with or without blocking indicates impaired neuromuscular transmission. SFEMG is abnormal in >95% of patients with generalized myasthenia gravis, even when RNS is normal. However, increased jitter is not specific to NMJ disorders and can also occur in neuropathies and myopathies with ongoing reinnervation.
References
- Preston DC, Shapiro BE. Electromyography and Neuromuscular Disorders: Clinical-Electrophysiologic-Ultrasound Correlations. 4th ed. Elsevier; 2021.
- Dumitru D, Amato AA, Zwarts MJ. Electrodiagnostic Medicine. 2nd ed. Hanley & Belfus; 2002.
- Jin PH. Localization and diagnostic evaluation of peripheral nerve disorders. Continuum (Minneap Minn). 2024;30(5):1236–1264.
- Kimura J. Electrodiagnosis in Diseases of Nerve and Muscle: Principles and Practice. 5th ed. Oxford University Press; 2023.
- England JD, Gronseth GS, Franklin G, et al. Practice parameter: evaluation of distal symmetric polyneuropathy: role of autonomic testing, nerve biopsy, and skin biopsy. Neurology. 2009;72(2):177–184.
- Van den Bergh PYK, van Doorn PA, Hadden RDM, et al. European Academy of Neurology/Peripheral Nerve Society guideline on diagnosis and treatment of chronic inflammatory demyelinating polyradiculoneuropathy. J Peripher Nerv Syst. 2021;26(3):242–268.
- Uncini A, Kuwabara S. Electrodiagnostic criteria for Guillain-Barré syndrome: a critical revision and the need for an update. Clin Neurophysiol. 2012;123(8):1487–1495.
- Olney RK. Consensus criteria for the diagnosis of partial conduction block. Muscle Nerve. 1999;22(S8):S225–S229.
- Cornblath DR, Asbury AK, Albers JW, et al. Research criteria for diagnosis of chronic inflammatory demyelinating polyneuropathy (CIDP). Neurology. 1991;41(5):617–618.
- Barohn RJ, Amato AA. Pattern recognition approach to neuropathy and neuronopathy. Neurol Clin. 2013;31(2):343–361.
- Rubin DI. Normal and abnormal spontaneous activity. Handb Clin Neurol. 2019;160:257–279.
- Stashuk DW. Decomposition and quantitative analysis of clinical electromyographic signals. Med Eng Phys. 1999;21(6–7):389–404.
- Wilbourn AJ. Nerve conduction studies: types, components, abnormalities, and value in localization. Neurol Clin. 2002;20(2):305–338.
- Gutmann L, Gutmann L. Axonal channelopathies: an evolving concept in the pathogenesis of peripheral nerve disorders. Neurology. 1996;47(1):18–21.
- Bodofsky EB, Carter GT, England JD. Is electrodiagnostic testing for polyneuropathy overutilized? Muscle Nerve. 2017;55(3):301–304.
- Tankisi H, Pugdahl K, Fuglsang-Frederiksen A, et al. Pathophysiology inferred from electrodiagnostic nerve tests and classification of polyneuropathies. Clin Neurophysiol. 2005;116(7):1571–1580.
- Daube JR, Rubin DI. Needle electromyography. Muscle Nerve. 2009;39(2):244–270.
- Stalberg E, Falck B, Sonoo M, Stalberg S, Astrom M. Multi-MUP EMG analysis — a two year experience in daily clinical work. Electroencephalogr Clin Neurophysiol. 1995;97(3):145–154.
- London ZN. A structured approach to the diagnosis of peripheral nervous system disorders. Continuum (Minneap Minn). 2020;26(5):1130–1160.
- Cartwright MS, Demar S, Griffin LP, et al. Validity and reliability of nerve and muscle ultrasound. Muscle Nerve. 2013;47(4):515–521.
