Nerve conduction studies (NCSs) and needle electromyography (EMG) are extremely useful in the diagnostic evaluation of the peripheral nervous system.1-7 They are a valuable adjunct to the clinical examination, localizing abnormalities along peripheral nerves or lower motor neurons.
Nerve pathology can occur anywhere along the course of the peripheral nerve—from the anterior horn cell, nerve root, plexus, peripheral nerve, or neuromuscular junction to the muscle. For this reason, these studies are useful in the evaluation of such conditions as motor neuron disease, radiculopathy, plexopathy, polyneuropathy, entrapment neuropathy, traumatic nerve injury, neuromuscular junction defect, and myopathy.
In order to optimize the diagnostic potential of these tests, a basic understanding of the neurophysiology as well as the methodology is needed.
Anatomic and Physiologic Basis for Electromyographic Studies
Motor Unit and Action Potential
The lower motor neuron nervous system consists of the anterior horn cells, the peripheral nerves, neuromuscular synapses, and muscles. The motor unit is comprised of the motor neuron, the axon, and all muscle fibers innervated by the motor neuron. Stimulation of the motor neuron causes contraction of all of the muscle fibers innervated by that neuron. NCSs and EMG evaluate the function of the motor unit. A motor unit potential (MUP) is the action potential created by the voluntary contraction of muscle in the motor unit and recorded by needle EMG.
Muscle and nerve cells have a unique characteristic not possessed by other cells of the body—excitability. Resting nerve and muscle cells have a normal negative electrical potential (ie, negative inside compared with the outside of the cell) called the resting membrane potential. When the cell is electrically stimulated, a brief depolarization occurs, resulting in a relative positive change in the membrane potential. If a threshold level of depolarization occurs, an action potential (AP) is generated (Fig. 1). The ability of an external electrical stimulus to produce depolarization and excitation depends on the intensity (ie, strength) and duration of the stimulus.1-3 The intensity and duration of an electrical stimulus to produce excitation of a nerve depends on nerve excitability. Injury or dysfunction of a nerve decreases nerve excitability; this finding can be one of the first electrophysiologic indicators of nerve injury. Because nerve excitability typically becomes abnormal 72 hours after a severe nerve injury, within 72 hours of a nerve injury, normal nerve conduction is preserved in the nerve segment distal to the site of nerve injury.8
Propagation of the Action Potential and Nerve Conduction Velocity
Once initiated, the AP is selfpropagating along the entire nerve cell membrane (Fig. 1). The speed of conduction is largely dependent on nerve fiber diameter and the presence or absence of myelin. Large-diameter fibers conduct faster than small-diameter fibers because of their lower resistance, and myelinated fibers conduct faster than unmyelinated fibers. In unmyelinated nerve fibers, depolarization of one segment of the nerve fiber generates an electrical current in the next segment. Because of this local circuit conduction, nerve conduction velocity (NCV) in unmyelinated fibers is relatively slow (1 meter per second). In myelinated fibers, the myelin sheath acts as an insulator and prevents transmembrane current flow between the nodes of Ranvier, resulting in saltatory conduction. The conduction velocities in these fibers can range from 3 to 80 m/s.6
Pathologic processes can affect the NCV.1-6 With segmental demyelination of the nerve, there is significant slowing of the NCV because of the prolongation of the AP rise time, secondary to leakage of the electrical current in demyelinated areas and loss of saltatory nerve conduction. With axonal degeneration in which the myelin —an anatomic substrate of nerve conduction—is relatively intact, the NCV remains the same or is only minimally affected (ie, slows no more than 40% below the normal mean value). This occurs because the remaining myelinated fibers conduct at the normal velocity. Only in axonal processes in which the large-diameter fibers are significantly affected is there prominent NCV slowing. Based on the speed of the NCV, the electromyographer often judges whether the patient has a demyelinating process or axonal degeneration.1-3 Focal demyelination is the basic pathologic process in focal entrapment. With entrapment neuropathies, such as carpal tunnel syndrome or ulnar nerve entrapment neuropathy at the elbow, there is focal slowing of the NCV.
Propagation of the AP results in generation of a compound nerve action potential (NAP), a summated potential of all nerve fibers stimulated by an electrical impulse. The NAP of motor nerves triggers the release of acetylcholine from the presynaptic nerve terminal. Acetylcholine diffuses into the synaptic cleft and binds to receptors on the postsynaptic membrane (Fig. 2). This binding induces depolarization of the muscle membrane and produces a nonpropagated and graded muscle end plate potential (EPP). If the EPP reaches threshold through simultaneous release and binding of a large number of acetylcholine quanta, a muscle action potential (MAP) is generated at the motor end plate (ie, neuromuscular junction) and the muscle fiber contracts.
Neuromuscular transmission can be inhibited in several ways.1-6 Presynaptic inhibition occurs in Lambert-Eaton myasthenic syndrome, in which the main defect is an insufficient release of acetylcholine because of an antibody directed against the voltage-gated calcium channel. This prevents the normal influx of calcium, which then results in reduced release of acetylcholine. Botulinum toxin inhibits the release of acetylcholine, producing a neuromuscular blockade. The most common neuromuscular transmission disorder is myasthenia gravis, in which an antibody binds to and alters the acetylcholine receptor site, causing a postsynaptic defect. The repetitive nerve stimulation test is used to diagnose neuromuscular transmission disorders. A low-rate (2 to 5 Hz) stimulation test will show a decremental response (ie, gradual decrease of the compound muscle action potential [CMAP] with repetitive nerve stimulation). A high-rate (10 to 50 Hz) stimulation test will distinguish between a presynaptic and postsynaptic disorder. A markedly abnormal incremental response is seen with Lambert-Eaton myasthenic syndrome, a classic presynaptic disorder.
When a muscle fiber is denervated, its sensitivity to acetylcholine increases greatly over a 1- to 2-week period. With this denervation hypersensitivity, the muscle fiber responds to a smaller than normal amount of acetylcholine. Fibrillation potentials, representing the depolarization of single muscle fibers, are thought to represent this denervation hypersensitivity. These potentials are usually detectable by needle EMG in the muscle within 2 weeks of muscle denervation.
Skeletal muscle consists of thousands of individual muscle fibers. Muscle fibers contain bundles of myofibrils, which are comprised of many filaments. Each myofibril is divided into units called sarcomeres. Thick myosin filaments are interposed between thin actin filaments. During muscle contraction, the actin filaments slide between the myosin filaments, thus shortening the sarcomere.
When neuromuscular transmission results in an EPP that reaches threshold, a MAP is generated. The MAP is propagated by local circuit current flow along the membrane at a relatively slow rate of approximately 5 m/s. The MAP spreads through the muscle by the transverse tubule system. This stimulates the release of calcium, which then initiates actinmyosin coupling and muscle contraction (Fig. 2).
Voluntary activation of the motor neuron results in contraction of all the muscle fibers innervated by the motor neuron. This activation produces a recordable MUP on the EMG. The amplitude of the MUP is proportional to the number of muscle fibers that contract with activation of one motor neuron; the duration of the MUP reveals the synchrony of firing of muscle fibers. The amplitude and duration of the MUP (Fig. 3) are important in differentiating various neuromuscular diseases, such as denervation processes and myopathy.1-3,5 In chronic denervation processes, surviving nerve fibers increase their motor unit territory and fiber density, producing high amplitude-long duration MUPs. In myopathy, the motor unit territory and fiber density decrease because of muscle fiber degeneration, producing small amplitude-short duration MUPs. Myotonia is a disorder of the muscle fiber membrane involving an abnormally low muscle fiber threshold of depolarization. It is difficult for the contracted muscle to relax. Electromyographically, highfrequency discharges wax and wane with a characteristic “dive bomber” sound.
Nerve Conduction Studies
Nerve conduction studies (motor, sensory, or mixed) evaluate the function of a particular nerve by electrically stimulating the nerve and recording the response either in the muscle or nerve.1-5 The response, which is compared with normative data based on the specific nerve and a specific site, can yield useful information about axonal loss and demyelination. The precise localization and pattern of involvement of the pathologic process can also reveal a specific diagnosis.
Motor Nerve Conduction
The motor nerve conduction test is performed by stimulating a peripheral nerve with a single supramaximal stimulus at each of at least two proximal points along the course of the nerve (Fig. 4). The resulting CMAP is recorded with a surface electrode from a muscle (eg, abductor pollicis brevis) innervated by the stimulated nerve (eg, median nerve). The latency, measured in milliseconds (ms), is the time required from nerve stimulation to production of the CMAP. Physiologically, this requires axonal transmission of the nerve impulse, neuromuscular transmission, and muscle fiber depolarization. The terminal, or distal, latency is the time (in ms) required for this response from the distal (versus proximal) point of stimulation. The conduction time (in ms) is obtained by subtracting the terminal latency from the latency derived from the proximal point of stimulation. This conduction time will measure only the time of axonal transmission because the times for neuromuscular transmission and muscle depolarization have been subtracted.
The NCV, which is measured in meters per second, is obtained by dividing the conduction time by the distance between the proximal and distal stimulating sites. The NCV represents the maximum conduction velocity of the fastest nerve fibers. For the NCV to be diminished, almost all of the nerve's motor fibers must be slowed. An affected motor nerve with only a few intact nerve fibers may yield a normal NCV. Therefore, the NCV is normal or minimally affected in axonal processes in which the remaining nerve fibers conduct at a normal rate. For example, in patients with carpal tunnel syndrome, one of the nerve conduction abnormalities is a prolonged distal latency because the carpal canal is located distal to the most distal site of stimulation of the median nerve. In general, the median nerve NCV calculated on the forearm segment (proximal to the carpal tunnel) is normal. However, mild slowing in this segment also can be seen in 11% to 35% of patients.9
In addition to the NCV, NCSs reveal evidence about axonal integrity and synchrony of conduction.1-5 The amplitude, duration, and shape of the CMAP are also analyzed. The CMAP amplitude is a rough estimation of the number of nerve fibers that are activated by nerve stimulation and, subsequently, the number of muscle fibers that contract. Thus, a low CMAP amplitude reveals evidence of severe impairment of nerve conduction that occurs with axonal loss. The CMAP duration yields information about the synchrony of conduction through the individual nerve fibers and contraction of the muscle fibers. A prolonged duration reveals marked slowing of the NCV in some of the nerve fibers, which occurs in demyelination. This is the basis of the dispersion phenomenon (Fig. 5), in which the CMAP is split into numerous phases, representing a wide range of conduction velocities. In some instances of demyelination, some fibers conduct so slowly that the signal is lost altogether. This can produce a marked reduction in the CMAP amplitude on proximal stimulation compared with distal stimulation below the area of focal demyelination. The marked reduction in CMAP amplitude is known as conduction block and is caused by failure of electrical nerve conduction across the area of dymelination. This conduction block occurs with ulnar entrapment neuropathy at the elbow. The marked slowing of the NCV, conduction block, and CMAP dispersion are typical of demyelination, whereas a normal or mildly slowed NCV with reduced CMAP amplitude is typical of axonal loss.
Sensory and Mixed Nerve Conduction
The sensory nerve conduction test is performed orthodromically (ie, distal to proximal) by stimulating a sensory nerve distally and recording a compound nerve action potential (CNAP) proximally. The test also can be performed antidromically (ie, proximal to distal) by stimulating a sensory or mixed nerve proximally and recording the CNAP distally in the nerve (Fig. 6). The latency and NCVs are identical with either testing method. The latency is a measure of the time required for conduction of the nerve impulse from the point of stimulation to the point of recording. Unlike motor nerve conduction, in the sensory nerve test the sensory nerve conduction time is equal to the latency. The NCV is determined by dividing the distance by the conduction time (latency) and represents the maximum conduction velocity of the fastest sensory fibers. Similar to motor nerve conduction tests, the NCV is most affected by demyelination. The amplitude and duration of the CNAP are also measured, providing a rough estimation of the number of large nerve fibers that are activated by nerve stimulation. The CNAP amplitude is most affected by axonal loss. Low amplitude usually represents severe impairment of nerve conduction. The CNAP duration is the range of NCVs of the various largediameter sensory nerve fibers. Prolonged temporal dispersion (ie, dispersion phenomenon) also occurs with demyelination but is seen only with the near-nerve needle recording technique using a needle electrode rather than a surface electrode.
The mixed NCS is performed by stimulating the distal part of the mixed nerve (sensory and motor fibers) and recording the CNAP directly over the proximal site of the nerve. The NCV and CNAP amplitude are determined with the same technique as the sensory nerve conduction test.
In general, the sensory and mixed NCSs are much more sensitive than motor nerve conduction for detecting neuropathy because the CNAP is smaller than the CMAP. In addition, with neuropathy, the sensory fibers are affected before the motor fibers. Therefore, when diagnosing carpal tunnel syndrome, the palm-wrist sensory nerve conduction is more sensitive than the motor distal latency.1-3
Two electrophysiologic studies, or late responses, are commonly used in the EMG laboratory: the H-reflex and F-wave. These late-response tests are used primarily to evaluate the function of the proximal segment of peripheral nerves, including the roots and anterior horn cells.
The H-reflex represents a monosynaptic reflex from the gastrocnemiussoleus muscles and is the electrophysiologic counterpart of ankle reflex. The reflex is elicited from the calf muscle by stimulation of the posterior tibial nerve. It measures the latency over the monosynaptic reflex arc through the afferent Ia fibers, a monosynaptic connection in the spinal cord, and the efferent alpha motor fibers of the S1 nerve root. With peripheral neuropathy or S1 radiculopathy, the H-reflex in the gastrocnemius-soleus muscle is often absent or prolonged in latency because the ankle reflex is frequently absent.
The F-wave is an antidromic volley evoked by the supramaximal stimulation of the distal nerve during the motor NCS. The F-wave is considered to be a recurrent discharge of a few motor neurons activated by antidromic volleys in the motor fibers. The F-wave has been used as a measure of the conduction velocity of proximal motor fibers and for a proximal nerve conduction abnormality, as in neurogenic thoracic outlet syndrome. The F-wave, although not diagnostic, is either absent or mildly prolonged in latency in severe cases of radiculopathy.
Normal Values of NCSs and Factors That Affect Results
Normal values of NCSs are different from one laboratory to another and are usually determined by 2 SD from the normal mean.2 Normal values depend on the specific technique used; thus, care should be taken to compare a given value to the normative values of the given technique (Table 1). A comparison of normal values from one laboratory to another is not possible unless the same technique is used by both laboratories. NCSs can be affected by many physiologic and technical variables.1-4 The most common factors are temperature, age, height, and distance measurements. If these factors are not controlled, the normal values will be inaccurate.
NCV increases linearly with increasing temperature. The skin temperature must be measured and adjusted to a set standard temperature with a temperature control unit or warm-water immersion. Temperature control is an important factor for the NCS and can be used as the quality control. If a laboratory does not measure skin temperature, the reliability of the results is suspect.
Nerve conduction in the full-term newborn is about 50% that of normal adult values, reaching about 75% of the adult value at 1 year and nearing 100% at 4 years of age.1 In adults older than age 60 years, the NCV may exhibit a gradual decline. The amplitudes of the CMAP and CNAP also show a gradual decline with age, especially after age 60. Normal values are usually based on adults between 18 and 60 years of age.
Height usually does not affect the NCV; however, there can be slight slowing of the NCV in very tall individuals. Height can also affect the results of the F-wave and H-reflex late-response studies. Because of the path of the nerve impulses through afferent nerve fibers, the spinal cord, and efferent pathways, which vary with height, the F-wave and H-reflex latency values are heavily dependent on body length (ie, height).
Probably the most common technical error made in performing NCSs is obtaining an accurate distance measurement. The distance is affected by limb positioning (eg, elbow flexionextension) and the contour of the anatomic part (eg, thoracic outlet segment). Proper technique, including placement of the electrodes and stimulator as well as increasing the stimulus to maximal stimulation, is also critical in obtaining accurate data.
Needle EMG is used to study the physiologic status of muscle function.1-7 With EMG, a needle electrode is used to measure the intrinsic electrical activity of muscle fibers. The needle EMG study has three components —observation at rest, MUP on minimal voluntary contraction, and the recruitment pattern of MUPs on maximal contraction. At rest, the muscle should be electromyographically quiet. Insertional activity is the electrical response of muscle membrane to insertion of the EMG needle. Increased insertional activity, or persistent abnormal spontaneous electrical activity, is secondary to a hyperexcitable muscle membrane. Decreased insertional activity can be associated with muscle fibrosis, fatty replacement of the muscle, muscle paralysis, or myopathy.
Several abnormal spontaneous potentials —activity that persists beyond the cessation of needle insertion—can be detected, including fibrillations, positive sharp waves (PSWs), fasciculations, myokymic potentials, complex repetitive discharges, and myotonic discharges. Fibrillations, PSWs, and fasciculations are commonly observed with denervation; however, fibrillations and PSWs can also be seen with active myopathy. Fasciculations and myokymic potentials typically occur with neurogenic lesions, whereas complex repetitive discharges and myotonic discharges typically occur with myopathies.2-5
Motor Unit Potentials
The MUP represents only the summated activity of the muscle fibers in a motor unit that are near the needle electrode, not the activity from all fibers within a motor unit. The MUP amplitude, duration, shape, and rate of firing are indicators of motor unit function.2,3,5 The amplitude primarily reflects the summated activity from fibers nearest the tip of the recording electrode. The duration depends on the depolarization of many muscle fibers away from and close to the needle tip. In normal individuals, the MUP amplitude and duration are relatively constant. A small amplitude-short duration MUP is typical of myopathy because many of the individual muscle fibers in a myopathy do not function properly. Therefore, the summation of the number of muscle fibers that activate with the activation of one motor unit decreases, producing a small MUP. With muscle denervation and collateral reinnervation (ie, reinnervation occurring from collateral sprouting from undamaged axons to adjacent, damaged axons), more muscle fibers are activated by the stimulation of one motor unit, resulting in high amplitude-short duration MUP (Fig. 3).
Needle EMG can also reveal other pathophysiological abnormalities.2-5 When the needle is inserted into a resting muscle, no persistent electrical discharge should occur. In both denervation and active myopathic processes, the muscle membrane is unstable and produces spontaneous discharges: fibrillation potentials and PSW potentials. Fibrillation potentials are low-amplitude, short biphasic or triphasic potentials (Fig. 7, A). PSWs are positive deflections followed by a prolonged negative wave (Fig. 7, B). In denervation, both fibrillations and PSWs can be seen within 2 to 3 weeks of muscle denervation. The time factor for the presence of PSW and fibrillation potentials is important.8 If the needle EMG test is performed during the first week after a nerve injury, PSWs and fibrillations in the denervated muscles are often absent. However, if PSWs and fibrillations are documented, the nerve injury is generally older than 8 days. Usually PSWs are seen earlier than fibrillations after a nerve injury. Fasciculation potentials are spontaneous discharges of a whole motor unit and can be seen in normal muscle and in anterior horn cell diseases, cervical spondylotic myelopathy, radiculopathy, and demyelinating neuropathy.
The rate and pattern of MUP firing, also known as recruitment, is also helpful in evaluating disease processes. In denervation processes, there is a reduction in the number of MUPs produced by the patient with muscle contraction, resulting in a reduced interference pattern. In myopathy, early recruitment of MUPs occurs—that is, more MUPs are present than would be expected for the degree of muscular contraction. Each MUP is inefficient because of fewer healthy, functioning muscle fibers. Thus, needle EMG is extremely useful in detecting myopathy. It is also useful in detecting denervation as well as determining its chronicity. Prominent fibrillations and PSWs are indicative of active denervation, and high amplitude-long duration MUPs are indicative of chronic denervation.
Factors Affecting Electromyographic Potentials
Several factors can influence the MUP, including the type of needle used (monopolar versus concentric), needle position, temperature, muscle type, and age.2-5 The MUP is higher and the duration longer with a monopolar needle than with a concentric one. Obtaining maximum MUP amplitude, a crisp and sharp sound with needle placement, and a short rise time is indicative of optimal needle placement. A change by 1 mm of the needle position from depolarizing fibers can decrease MUP amplitude by 90%. MUP duration and amplitude increase with decreasing temperatures. Smaller muscles, such as ocular muscles, have a shorter MUP duration than larger muscles, such as the quadriceps. The MUP duration and amplitude increase with age. Patient cooperation and an accurate assessment of the patient's clinical problem are crucial.
Peripheral Nerve Entrapment
In peripheral nerve entrapments, such as carpal tunnel syndrome (CTS)1-5,7,10-12 and cubital tunnel syndrome (ulnar neuropathy at the elbow),1-5,13,14 NCSs are very useful in confirming the diagnosis. External neural pressure initially results in focal demyelination. Therefore, NCSs reveal marked NCV slowing across the nerve segment being compressed. The CMAP may also reveal dispersion and conduction block across this site. Although both motor and sensory fibers are affected, in most patients the sensory nerve conduction is affected first. The more sensitive NCS can isolate the area of compression.
If the nerve compression is severe, axonal loss occurs. In such cases, the NCS reveals a loss of CNAP and CMAP amplitude. Axonal loss can also result in an abnormal EMG. Muscles innervated by the affected nerve frequently demonstrate fibrillation and PSW potentials, reduced recruitment, and giant MUPs if subsequent collateral reinnervation of the muscle occurs. The degree of NCV slowing and the presence of secondary axonal changes are helpful in grading the severity of the nerve compression. These studies can also be performed serially to evaluate the severity of the compression.
Carpal Tunnel Syndrome
The test for CTS should include motor, sensory, and mixed NCSs in the median and ulnar nerves in the symptomatic arm. If CTS is confirmed in the symptomatic arm, the terminal latency and sensory conduction should be checked in the opposite median nerve. A needle EMG in the abductor pollicis brevis muscle is recommended to document any secondary axonal degeneration.2
The most sensitive CTS study is the sensory nerve conduction test across the palm-to-wrist segment. Other sensitive tests include the finger-to-wrist NCS and the distal motor latency. The diagnosis of CTS can be confirmed by the NCS with a sensitivity (true positive) in 85% to 91% of cases and a specificity (true negative) in 95% of cases.1,12 In a small number of patients, a comparative sensory NCS is needed, such as the ulnar-median sensory nerve conduction comparison with stimulation of the ring finger (the ring finger has median and ulnar nerve dual innervation) or the radial-median sensory nerve conduction comparison with stimulation of the thumb (the thumb has superficial radial and median nerve dual innervation).10
A 42-year-old computer programmer presents with a 3-month history of right hand pain and numbness. The pain radiates into all the fingers. The examination is normal except for a Tinel sign at the right wrist and decreased two-point and Semmes-Weinstein monofilament sensation in all fingers of the right hand. CTS is clinically suspected, but polyneuropathy is also a consideration because the numbness involves the ulnar innervated areas as well. The NCS reveals marked prolongation of the right median terminal latency (5.7 ms) and a slow NCV in the palm-to-wrist segment of the median sensory nerve (25 m/s). The CMAP amplitude is normal, but the CNAP amplitude of the sensory segment is decreased (Fig. 8). The proximal segments of the median nerve and all segments of the right ulnar nerve and radial sensory nerve are normal. The EMG of the abductor pollicis brevis muscle (median nerve innervated) and first dorsal interosseous muscle (ulnar nerve innervated) are normal. These findings confirm a demyelinating process across the palm-to-wrist segment of the median nerve and are indicative of CTS. The normal distal NCS in the distal ulnar and radial nerve exclude polyneuropathy.
Cubital Tunnel Syndrome
The tests for cubital tunnel syndrome should include motor, sensory, and mixed NCSs in the median and ulnar nerves in the symptomatic arm. Special attention should be paid to the motor NCV across the elbow, as this segment isolates the area of entrapment and aids in the precise localization of the site of nerve compression. The sensory and mixed NCSs from finger-to-wrist and wristto-elbow segments are frequently abnormal, as well. An inching technique of the ulnar motor nerve is helpful in localizing the site of nerve compression. In this technique, a 10-cm segment across the elbow is divided into 2-cm segments beginning 6 cm below the medial epicondyle and ending 4 cm or more proximal to it. The NCSs enable localization of an ulnar nerve compressive lesion in 82.5% to 96% of cases.14
A 58-year-old man presents with numbness in the ring and little fingers and weakness of intrinsic muscles in his hand. Ulnar neuropathy is clinically suspected. Ulnar motor NCSs reveal marked slowing of the NCV across the 10-cm elbow segment (25.6 m/s) and conduction block (CMAP amplitude 13.1 mV in the lower arm segment and 1.8 mV across the elbow). Sensory and mixed NCSs reveal no potential from the elbow segment distally. Median NCS is normal. These results indicate ulnar neuropathy across the elbow. To localize the area of entrapment more precisely, a motor nerve inching technique is performed by stimulating each site along the nerve in 2-cm segments (Fig. 9). The most prominent latency difference in this study occurs over the segment starting 2 cm below the medial epicondyle to the level of the medial epicondyle (latency difference, 2.0 ms). Conduction block is also seen across this segment. This indicates focal demyelination at this site.
In radiculopathy, the most important study is the EMG.2,15 The needle EMG should include both the paraspinal and limb muscles. With nerve root compression, the sensory NCSs are normal. This occurs because the sensory cell body, the dorsal root ganglion, is distal to the site of nerve compression. There is no injury between the cell body and axon terminal; therefore, the sensory NCS will be normal. With a motor NCS, a pure demyelinating lesion will not be seen because the area of injury (and focal area of slowing) is proximal to the tested nerve sites. If the nerve root compression is severe enough to result in axonal injury and loss, the motor nerve study will reveal decreased CMAP amplitudes in all tested segments of the motor nerve.
If axonal loss does occur, the EMG examination will reveal denervation changes in muscles innervated by the specific nerve root. The radicular pattern of muscle involvement leads to a specific root level. The paraspinal muscles must also be evaluated. If the paraspinal muscles are involved, radiculopathy is confirmed. However, a negative EMG study does not exclude a radiculopathy, particularly if only the sensory portion of the nerve is involved or there is no motor axonal loss. The sensitivity of NCS/EMG in radiculopathy is approximately 80% to 85%.2 Some muscles do not demonstrate clear denervation changes on EMG until up to 3 weeks after an axonal injury. Consequently, a study performed too soon after an acute injury may be misleading.
NCSs are extremely helpful in defining the severity and distribution of nerve injury in cases of trauma.8,16,17 Immediately after a total nerve transection injury, no motor or sensory response will occur across the injured segment, nor will there be any voluntary MUPs from muscles innervated by that nerve. Diminished nerve excitability will be exhibited in 1 to 3 days, along with a gradual re duction in the CMAP and CNAP amplitudes in nerves stimulated distally to the region of injury. About 4 to 5 days after injury, absence of CMAP amplitude exists in the motor NCS as well as further reduction in CNAP amplitudes in the sensory NCS. After 6 to 10 days, there is no CNAP response. PSW potentials are seen at 8 days after nerve injury and fibrillation potentials at 14 days. This indicates that the complete extent and severity of injury can be assessed adequately only after the third week. In nerves that are severely injured but not severed, proportional reduction in CMAP and CNAP amplitudes occurs, and needle EMG reveals voluntary MUP in muscles innervated by the injured nerve. In some cases of severe but incomplete nerve injury, initially there are no obtainable voluntary MUPs on needle EMG, usually because of severe nerve injury producing a conduction block. In these cases, a repeat study in 6 to 8 weeks can demonstrate a few smallamplitude polyphasic MUPs indicative of reinnervation. This repeat study can therefore be used to determine whether surgical exploration is warranted.
In polyneuropathy, NCSs define the pathologic process, extent of involvement, and severity of the disease.2,3,6 NCSs can reveal evidence of subclinical polyneuropathy in patients presenting with more focal symptoms, which is extremely important when surgical intervention is being considered for entrapment neuropathies with superimposed polyneuropathies. Additionally, the findings of demyelination and axonal loss are critical in the diagnosis of acquired and hereditary polyneuropathies. Specific diagnoses, such as Charcot-Marie-Tooth disease and chronic inflammatory demyelinating polyneuropathy, can be made on the basis of these tests. If there is a suspicion of polyneuropathy, NCSs should be performed in at least three limbs. Even in the patient with suspected CTS, other distal nerves, such as the ulnar and radial nerves, should be evaluated in that limb in addition to the median nerve, to look for subclinical polyneuropathy.
A 55-year-old man with a 15-year history of diabetes mellitus presents with left-hand numbness. His examination reveals no strength limitations. He has numbness to pinprick testing and decreased two-point and Semmes-Weinstein monofilament sensation in all fingers of the left hand. He also has pinprick loss in both legs up to the knee. The focal nature of his symptoms suggests nerve entrapment such as CTS, but diffuse findings on examination suggest diabetic polyneuropathy.
NCSs reveal decreased CNAP amplitudes in left median, ulnar, and radial sensory nerves (3 μV, 4 μV, and 3 μV, respectively) and mild to moderate NCV slowing (38 m/s, 36 m/s, and 40 m/s, respectively). Proximal motor and mixed NCSs of the median and ulnar nerves from the axillaryto-wrist segment are normal. NCSs of the left leg reveal absent sural nerve potentials and reduced CMAP amplitudes with mildly slowed NCV (35 to 36 m/s) in the peroneal and posterior tibial motor nerves. EMG examination reveals fibrillation and PSW potentials in the left tibialis anterior and gastrocnemius muscles, with reduced recruitment on maximal muscle contraction. These electrophysiologic results are indicative of polyneuropathy. The reduced CMAP and CNAP amplitudes with mild to moderate NCV slowing in the distal segments of the nerves are very suggestive of axonal loss. In polyneuropathy, distal nerve conduction and EMG abnormalities are typical, with the distal segments being affected initially. In some cases of axonal polyneuropathy, there is evidence of superimposed nerve entrapment. NCSs are critical in these cases and demonstrate focal demyelination superimposed over a diffuse polyneuropathy.
Understanding the neurophysiology and methodology of electrophysiologic studies allows clinicians to maximize their diagnostic capabilities. NCSs and EMG are extremely useful in confirming and localizing suspected entrapment neuropathy, radiculopathy, and polyneuropathy, and in defining the severity and distribution of injury in nerve trauma. These studies can provide support for the consideration of a surgical procedure and provide an objective method for serial evaluation of the status of the nerve.