Electrodiagnosis is a broad term that includes multiple electrodiagnostic techniques, including needle electrode examination (NEE); motor and sensory nerve conduction studies (NCSs), including late responses; and evoked potentials (EPs). Electrodiagnostic techniques such as electromyography (EMG) and EPs are very useful adjuncts to physical examination of patients in pain. Conditions in which EMG or EPs may be of use include painful peripheral neuropathies, entrapment neuropathies, traumatic nerve injuries, radicular and multiradicular problems, lumbar spinal stenosis, arachnoiditis, and painful myopathies.
In recent years there has been increased reliance on anatomic measures such as intradermal nerve biopsy for small-fiber neuropathy and magnetic resonance imaging (MRI) for larger neural structures. This “anatomic” approach fails to recognize the differences and respective advantages and shortcomings of physiologic versus anatomic testing and, above all, fails to recognize the inherent nature of pain, a subjective experience that can only be correlated with the clinical picture.
The problems inherent in applying electrodiagnostic techniques to the diagnosis and management of pain are no different from those encountered in history taking, physical examination, radiologic evaluation, and therapeutic diagnostic testing (e.g., nerve blocks). Pain is a subjective experience, often without an objective “litmus test,” and final diagnosis of the etiology and presumptive treatment of a pain syndrome is a clinical one that can be supported only by relevant data, including findings on EMG.
Electrodiagnostic techniques can be applied virtually without complications to large portions of the body and hence the nervous system to gain an overall understanding of the distribution of abnormalities, if present, or relative normalcy if not. The distribution of abnormalities correlates strongly with the etiology of a disease process causing pain. Questions that can be posed and answered through the use of electrodiagnostic techniques include the following:
- •
Is a disease of nerves present?
- •
If so, is it a mononeuropathy, polyneuropathy, or mononeuritis multiplex?
- •
Does the distribution of abnormalities suggest involvement at the nerve root, plexus, or nerve level?
- •
If there is a disease process involving a single (or multiple) nerve such as injury or compression, is it improving, worsening, or static?
- •
Is the nerve involved motor, sensory, or mixed?
- •
Is the process one of nerve, muscle, or both?
- •
Are small fibers selectively involved or is this mainly a large-fiber disorder, or are both involved?
- •
Is there autonomic as well as somatic involvement?
- •
Are more proximal structures rather than distal ones involved?
- •
Is the central nervous system involved?
These are the types of inquiries that add substantially to diagnostic information about the etiology, severity, and prognosis of painful disorders. Although many painful disorders do not affect either the peripheral or central nervous system, electrodiagnostic testing can often add reassuring negative results to the diagnostic picture.
Electrodiagnosis is an extremely useful investigative technique for evaluating patients with pain because it satisfies two fundamental steps in the assessment of a neuropathic painful syndrome before any attempt at therapy: (1) rigorously establishing the presence or absence of a peripheral nervous system lesion and (2) determining the relevance of an established peripheral neuropathic lesion to the subjective clinical complaint. In addition, with the advent of new treatments, including enzyme replacement and potential insertion of genetic material into cells, early diagnosis and the ability to monitor treatment outcomes objectively become of great importance. Recent enzyme replacement therapy for Pompe’s disease is an example of the potential diagnostic, therapeutic, and prognostic importance of electrodiagnosis.
The parent organization of electrodiagnosis is the American Association of Neuromuscular and Electrodiagnostic Medicine (AANEM), which has promulgated a Recommended Policy for Electrodiagnostic Medicine Position Statement . This document seeks to lay out the scope of electrodiagnosis by using consensus expert opinion to ensure adequate examination while attempting to conserve scarce resources by listing the indications for testing and the maximum number of studies per diagnosis in 90% of cases. Since there has been some abuse of the number of studies done on patients and since many pain specialists are not familiar with the techniques of electrodiagnosis, this may serve as an important source of information on the appropriate measures to be used when these diagnostic methods are being considered.
Electromyography
EMG is a method of testing both the physiologic state and the anatomic integrity of lower motor neuron structures (anterior horn cells, nerve roots, plexuses, peripheral nerves, neuromuscular junction, and muscles), their sensory components, and some spinal and brainstem reflex pathways. The term electromyography previously caused considerable confusion because strictly speaking, it was needle electrode evaluation (NEE) of muscle function but often was expanded to include nerve conduction velocity (NCV) or NCSs and other tests. However, its common usage has come to mean needle EMG, determination of NCV, and less frequently, testing such as the H-reflex and F-response , cranial nerve reflexes (e.g., the blink reflex), and studies of the neuromuscular junction. The all-inclusive term EMG is used here in an effort to avoid confusion among tests.
Longmire pointed out the puzzling dichotomy regarding pain and electrodiagnostic testing in the medical literature. In standard textbooks on electrodiagnostic testing, little or no reference is made to painful syndromes and their diagnosis despite their excellent discussion of physiology, technique, and clinical correlation. On the other hand, a perusal of pain textbooks reveals cogent attempts to correlate neurophysiologic studies with pain management. The reason for the paucity of references appears to be at least in part the attitudes of pain specialists themselves, who point out that “large caliber afferent fibers are physiologically unrelated to pain, a submodality mediated by small caliber fibers. Additionally, the test is unable to explore the bases for positive sensory phenomena, generated by dysfunction even of large caliber afferent channels.”
Electrical Testing of Nerves and Muscles
Brief History
As early as 1791, Galvani found that electricity was produced by muscular contractions. The first experimental work with EMG was performed by Lord Adrian in 1925. In 1928, Proebster first described the presence of “spontaneous irregular action potentials in denervated muscle.” In its progression to clinical application, EMG made a major step forward with use of the cathode ray oscilloscope by Erlanger and Gasser, as well as the concentric needle electrode and loudspeaker. Vast numbers of nerve injuries in World War II and later conflicts added further impetus to the study of nerve and muscle by electrodiagnostic technique.
The Electrodiagnostic Method
Electrical diagnostic measurement systems have only four basic components :
- 1.
Electrodes
- 2.
Stimulator
- 3.
High-gain differential amplifier
- 4.
Recording display or central processing device
The EMG apparatus amplifies and displays biologic information derived from either surface or needle electrodes. Electrical information may be recorded from muscles, nerves, or other nervous system structures and is displayed on an oscilloscope. In addition to the visual display on the oscilloscope, a permanent recording may be made, audio amplification may allow it to be heard over a loudspeaker, and analog-digital analysis of signals may be used. Nerves are electrically stimulated to measure conduction.
For NCSs, skin surface electrodes are generally used for recording compound muscle or nerve action potentials. Rarely, needle electrodes are used. For needle EMG, needle electrodes are used with a strong trend toward disposable needles. For sensory testing, ring electrodes are used for measurement ( Fig. 14.1 ). Modern EMG equipment is manufactured by numerous companies and is generally standardized to allow reliable and reproducible testing by different laboratories, but normative data, including data for special populations such as geriatric, pediatric, diabetic, and even active workers, may differ among laboratories and require standardization by each laboratory.
NEE is an invasive procedure but complications are rare; however, patients should be apprised of them. The most common is transient muscle soreness. Aseptic precautions should be observed. Precautions for testing include extra care with patients who are taking warfarin or other anticoagulants or who have hemophilia or other blood dyscrasias, but since most muscles tested are superficial, they can easily be compressed and the bleeding abated. Severe thrombocytopenia is a relative contraindication and should be considered carefully. Patients positive for human immunodeficiency virus (HIV) represent a transmission risk, but the use of disposable needles (which should be universal) should protect against this hazard. One should carefully consider persons who have a cardiac pacemaker or transcutaneous stimulator when doing stimulation for NCSs. Certain muscles, such as the rhomboids and abdominals, which are sometimes interrogated by NEE in pain patients, and the diaphragm (almost never), carry a risk for pneumothorax and infectious peritonitis, respectively. Beyond placing a needle through an infected site, there are probably no absolute but only relative contraindications to EMG. Extremely anxious patients and some children occasionally require some sedation. Aftereffects are negligible, with rare bruising, although occasionally a highly disturbed, suggestible, or litigious patient may complain vehemently of increased pain or disability. To the contrary, patients may occasionally report a salutary effect on their condition!
Physiology
Physiologic Mechanisms in The Production of Muscle Potentials
When an impulse arrives at the region of the junction between a nerve and muscle, at the nerve terminal, a depolarization takes place and triggers the opening of voltage-gated calcium channels, which in turn triggers the release of acetylcholine in the synaptic cleft between the axon terminal and the neuromuscular junction. Acetylcholine activates nicotinic receptors, which are ligand-gated sodium channels that activate the tubule system of the muscle. The stimulus is transmitted along the fiber by an excitable membrane that surrounds the muscle fiber. The action potential results from breakdown of the surface membrane potential, which is associated with critical changes in ionic permeability. In a resting muscle fiber, the potential difference across the surface membrane is 90 mV, with negative inside and positive outside. During excitation, the resting potential temporarily reverses to 40 mV, negative outside. This action potential travels along the muscle fiber at velocities ranging from 3.5 to 5 m/sec in different fibers.
In recording extracellularly, as with EMG, the electrode picks up the action potential as it is conducted through the medium that surrounds the active fiber. The impedance of the external medium is small in comparison to the impedance of the fiber interior, and hence the voltage of the extracellularly recorded potentials is maximally only 2% to 10% of the intracellularly recorded potential changes. The functional unit ( Fig. 14.2 ) in reflex or voluntary activity is the motor unit ; a motor unit is the group of muscle fibers innervated by a single anterior horn cell.
Conduction along the fine intramuscular branches of the anterior horn cell axon occurs so rapidly that all muscle fibers in a motor unit are activated nearly simultaneously. The number of muscle fibers per motor unit varies considerably from muscle to muscle; for example, in the gastrocnemius the motor unit consists of about 1600 muscle fibers, whereas in the small muscles of the eye there are only 5 to 10 fibers. The motor units in various muscles cover different areas of the muscle’s cross section (e.g., brachial biceps, 55 mm; rectus femoris, anterior tibial, and opponens pollicis, 8 to 9 mm). The distribution of fibers is such that fibers from several different motor units are intermingled, which is why four to six motor units can be identified by EMG from the same intramuscular recording point. In normal muscle, these single motor unit potentials can be differentiated only during weak voluntary effort.
The potentials from different motor units are recognized by their frequency of discharge, which varies for each motor unit (some are more or less excitable). Moreover, the various potentials often differ in appearance because of the differential distance of the recording electrode from the individual fibers of the activated motor units and the differential distribution of the motor end plates in the several units within “range” of a concentric or single needle electrode in one position in the muscle. An upward deflection on the oscilloscope is considered electrically negative, and a downward deflection is considered electrically positive. In the immediate vicinity of a potential there is an upward, or negative, deflection.
Physiology of Nerve Conduction
The cell membrane (axolemma) of a nerve axon separates the intracellular axoplasm from the extracellular fluid. The unequal distribution of ions between these fluids produces a difference in potential across the cell membrane. This resting potential is about 70 mV and is negative on the inside with respect to the outside of the cell membrane. When a nerve fiber is stimulated, it causes a change in the membrane potential; a rapid but brief flow of sodium ions occurs through ionic channels inward across the cell membrane and gives rise to an action potential.
The way in which an action potential is conducted along an axon depends on whether the axon is myelinated or unmyelinated. In a myelinated fiber, the action potential is regenerated only at the nodes of Ranvier, so the resulting action potentials “jump” from node to node, thereby resulting in saltatory conduction. The velocity of nerve conduction depends on the diameter of the myelinated fiber. Small myelinated fibers may conduct as slowly as 12 m/sec, whereas large motor and sensory fibers conduct at a rate of 50 to 70 m/sec in humans. In an unmyelinated fiber in humans, the conduction rate is about 2 m/sec.
Several factors affect conduction velocity other than whether the axon is myelinated:
- •
Temperature of the limb (low temperatures decrease conduction velocity)
- •
Age of the patient (infants have slow conduction velocities and older adults have increasingly slowed conduction velocities)
- •
Height of an individual (increased height may increase the internodal distances of the nodes of Ranvier)
Basic Electromyography Examination
EMG must be combined with clinical examination of the patient by the electromyographer. This includes grading of muscle strength. It is of prime importance for the electromyographer to personally correlate the clinical data and that obtained by EMG. Each examination must be planned individually. There is no “cookbook” formula to follow. Because the EMG examination is an extension of the clinical examination, the patient must be evaluated fully and the problem tentatively assigned to the portion of the anterior horn cell system that seems most likely to be involved. The electromyographer determines the segment or segments of the peripheral nervous system suspected to be involved, and the examination is planned to either substantiate or invalidate the presumptive clinical diagnoses.
Conducting The Examination
Needle examination of a patient is designed to determine the following:
- 1.
Integrity of a muscle and its nerve supply
- 2.
Location of any abnormality
- 3.
Any abnormalities of the muscle itself
The electrodes may be monopolar or concentric ( Fig. 14.3 ). The examination proceeds through the following steps ( Fig. 14.4 ):
- 1.
Determination of activity of the muscle in the relaxed state
- 2.
Evaluation of any insertional activity that arises
- 3.
Assessment of the activity seen on weak voluntary effort
- 4.
Determination of the pattern seen on maximum voluntary effort, which is known as the interference pattern (there is interference in discerning individual muscle action potentials from the resting baseline)
Needle Findings in Normal Muscle
Insertional Activity
When the needle is inserted into a normal muscle, it evokes a brief burst of electrical activity that lasts no more than 2 to 3 msec, a little longer than the actual movement of the needle. This activity is described as insertional activity and is generally 50 to 250 mV in amplitude (see Fig. 14.4 A ). These insertional potentials are believed to represent discharges from muscle fibers produced by injury, mechanical stimulation, or irritation of the muscle fibers.
Spontaneous Activity (Activity At Rest)
When the needle is stationary and the muscle is relaxed, there is no electrical activity present in normal muscle except when the needle is in the area of the end plate. Two types of end-plate “noise” are normal (see Fig. 14.4 C ): (1) low-amplitude and undulating, which probably represents extracellularly recorded miniature end-plate potentials, and (2) high-amplitude intermittent spike discharges, which probably represent discharges of single muscle fibers excited by intramuscular nerve terminals irritated by the needle. Any other spontaneous activity at rest is abnormal. An increased duration of insertional activity may be seen with loss of innervation or with primary disease of muscle fiber. Reduction may occur in patients with myopathies or more advanced degeneration in which muscle tissue has been replaced by fat or fibrous connective tissue.
Voluntary Activity
Voluntary activity of the muscle is analyzed after the muscle is studied at rest (see Fig. 14.4 D ). Electrical activity (termed a motor unit action potential ) is noted. As mentioned previously in the discussion on physiology, a motor unit refers to the number of muscle fibers supplied by one motor neuron and its axon. This number varies from muscle to muscle and may be as few as 10 to more than 1000 muscle fibers. When a motor neuron discharges, it activates all the muscle fibers of the motor unit.
The force of contraction determines the number of motor units brought into play. This begins with a single motor unit that fires and can be identified on the screen by its distinctive morphology. As effort is increased, other motor units come into play, which can still be individually discerned and have their own individual morphology and audio representation on the loudspeaker. As the contraction increases, the firing rate of each individual motor unit action potential increases, and the action potential is subsequently joined by other motor unit action potentials, whose firing rates also increase. This phenomenon is known as recruitment ( Fig. 14.5 ). In normal muscles, the strength of a voluntary muscle contraction is directly related to the number of individual motor units that have been recruited and their firing rate. Analysis of motor units includes (1) waveform, (2) amplitude, and (3) interference patterns.
Waveform
Most units are biphasic or triphasic. The number of phases is determined by the “baseline crossings.” Motor units that cross the baseline or have more than five phases are called polyphasic motor units . Though occasionally seen in healthy muscle, they do not exceed 15% of the total number of motor units. In some muscles, polyphasic motor units are more prevalent. Polyphasic potentials are a measure of fiber synchrony.
Amplitude
The amplitude depends on the number of fibers in the motor unit and the type of EMG needle used. Monopolar needles are associated with higher-amplitude potentials than bipolar or coaxial needles are. Normal amplitude ranges from 1 to 5 mV. Because the motor unit is the sum of the action potentials of each muscle fiber of the unit, a large motor unit has a larger amplitude; conversely, a smaller motor unit has a smaller amplitude.
Interference Patterns
With maximum voluntary effort, a large number of motor units are brought into play and their firing rate increases. They tend to “interfere” with each other and are not recognized further as individual units. This gives rise to a situation called an interference pattern (see Fig. 14.5 ). A normal muscle has a “full” interference pattern.
Needle Findings in Abnormal Muscle
Various abnormalities may occur that indicate the presence of total denervation—neurogenic paresis, peripheral type, or neurogenic paresis, anterior horn cell type. In addition, myogenic paresis may be detected. Generally, on the basis of abnormal findings and with a well-determined examination, the presence of a radiculopathy, generalized neuropathy, focal neuropathy or mononeuropathy, or plexopathy can be determined. The following are needle abnormalities in abnormal muscle:
- 1.
Insertional activity (decreased or increased)
- 2.
Spontaneous activity (fibrillations, positive sharp waves, or fasciculations) ( Fig. 14.6 )
Figure 14.6
Positive sharp wave and fibrillation potentials recorded from a denervated muscle.
- 3.
Abnormalities in voluntary motor unit activity, especially recruitment (see Fig. 14.5 )
- 4.
Abnormal motor unit morphology (e.g., excessive or extreme polyphasia)
Nerve conduction studies are of value in the following cases:
- 1.
Determining whether a disease of nerve is present
- 2.
Determining the distribution of a neuropathy (e.g., mononeuritis, polyneuropathy, mononeuritis multiplex; this may be a valuable point in the differential diagnosis of the cause of a neuropathy)
- 3.
Determining at what point in a nerve a conduction block is present and locating an entrapment site
- 4.
Studying the progress of disease of a peripheral nerve (e.g., Is it getting better? Worse? Staying the same?)
- 5.
Seeing whether reinnervation of a previously sectioned nerve has taken place
- 6.
Establishing in a disease of the myoneural junction (e.g., myasthenia gravis) the fact that conduction along the nerve is adequate or normal
There are a few important limitations to the use of NEE in muscle disease. Some disorders may not result in abnormal findings on EMG, including certain congenital conditions and endocrine disorders such as steroid myopathy and polymyalgia rheumatica, an important cause of muscle pain and weakness in the elderly. EMG affects primarily type I fibers during minimal voluntary effort before the interference pattern makes discernment of individual motor units difficult.
Conduction velocity studies are carried out by inserting a needle electrode into a muscle innervated by the nerve under study or by the use of surface electrodes over that muscle ( Fig. 14.7 ). For example, the first dorsal interosseous muscle may be examined to determine the function of the ulnar nerve ( Fig. 14.8 ). The nerve is stimulated at the elbow in the case of the ulnar nerve, and the latency of the response is determined. The response is generally a spike-like large motor unit action potential. The ulnar nerve is then stimulated in the wrist or the axillary region, or both. The difference in latencies between the two points of stimulation and the distance between the two points of stimulation provide the basis for calculation of conduction velocity. Conduction velocity is determined by the following formula:
MCV ( m / sec ) = DMM / ( PML − DML )
Textbooks of stimulation points and pickup points are readily available. Normal values are usually established for each nerve in individual laboratories, but normal values for commonly tested sensory and motor nerves are generally available ( Table 14.1 ). Median nerve stimulation is comparable to ulnar nerve stimulation ( Fig. 14.9 ).
| Nerves | Amplitude (Avg.) | Distal Latency in msec (Avg.) ∗ | Conduction Velocity in m/sec (Avg.) |
|---|---|---|---|
| Median (sensory) | 10-85 µV (20) | 2.0-3.7 (3.2) | |
| Ulnar (sensory) | 5-70 µV (15) | 1.6-3.2 (2.8) | |
| Radial (sensory) | 10-60 µV (18) | 1.7-2.8 (2.4) | |
| Median (motor) | 5-25 mV (8) | 2.0-4.0 (3.3) | 48-69 (54) |
| Ulnar (motor) | 5.5-20 mV (8) | 1.6-3.1 (2.6) | 50-69 (55) |
| Sural (sensory) | 3-38 mV (8) | 2.3-4.6 (4.1) | 41-61 (46) |
| Peroneal (motor) | 2.5-18 mV (4) | 2.3-6.0 (4.1) | 41-58 (45) |
| Posterior tibial (motor) | 4-38 mV (11) | 2.1-6.0 (4.3) |
∗ Distal latencies are based on standard distance: 13 cm for the median (S), 11 cm for the ulnar (S), 10 cm for the radial (S), 14 cm for the sural (S), 4 to 6 cm for the median (M) and ulnar (M), 6 to 8 cm for the peroneal (M), and 8 to 12 cm for the posterior tibial (M) nerves.
F Wave
Definition
Motor conduction velocity along the whole axon, including the proximal portions, can be studied by eliciting the F-wave response, a small, late muscle response that occurs as a result of backfiring of anterior horn cells. F waves may be obtained from almost any mixed nerve that can be stimulated, but the median, ulnar, peroneal, and posterior tibial nerves are the most commonly used ( Fig. 14.10 ). If the standard distal motor conduction velocities are normal but the F-wave value is prolonged, slowing must be occurring somewhere more proximal to the distal normal segment. (The method used to determine F-wave latency varies from laboratory to laboratory; the F-wave value with each successive shock stimulus shows a variability of several milliseconds, with some examiners averaging 10, 30, or 50 responses and some taking the shortest of 10, 20, or more responses.) Limb temperature and arm or leg length may also be important to know. Comparison with the opposite limb may be most helpful if that limb is asymptomatic.
Pitfalls and Comments
In addition to the variability in F waves and how they are obtained in different laboratories, many electromyographers overuse (or at least overperform) the F-wave study when proximal slowing in a nerve or nerve root is not even in the differential diagnosis. The most accepted use of the study is for suspected early Guillain-Barré syndrome , when results of the usual studies are still normal—typically in the first 10 days of the illness. It is highly controversial in the evaluation of radiculopathies.
Hoffman Reflex (H-Reflex)
Definition
The H-reflex is obtained by electrostimulation of the posterior tibial nerve in the popliteal space at a slow rate with a long duration and submaximal electrical shock; it is recorded with surface electrodes over the gastrocnemius-soleus ( Fig. 14.11 ). The impulse travels up the sensory fibers to the spinal cord, synapses with the alpha motor neuron, and returns down the motor fibers to the calf muscle. H-reflex latencies are therefore long, in the range of 40 to 45 msec. They are carried mostly in the S1 nerve root distribution and cannot be recorded consistently from other muscles. To determine a delay or an asymmetry, one should always study the opposite leg for comparison.
Pitfalls and Comments
The H-reflex is somewhat more useful than the F wave, but the main reason for the study is to evaluate patients with suspected S1 radiculopathy whose history or findings on physical examination are suggestive but the EMG is normal. Usually, when an absent H-reflex is noted, which suggests a problem with S1 nerve root conduction, an absent or depressed ankle reflex has already been noted on the physical examination, so the study is, for many, redundant. Pitfalls occur when the opposite leg is not studied to show a normal H-reflex as a contrast. If the H-reflex is absent bilaterally, it may reflect more generalized disease, such as peripheral neuropathy. Older patients often do not have good H-reflexes as a normal finding. In addition, a unilaterally absent H-reflex with normal findings on needle EMG does not indicate when the injury occurred; the findings may have been the result of a previous injury.
Quantitative Sensory Testing (Pseudomotor Axon Reflex Test)
Quantitative sensory testing takes various forms, including the quantitative somatosensory thermotest using a controlled ramp of ascending or descending temperature through a Peltier device. Measurement of the threshold for cold sensation reflects the function of small-caliber Aδ myelinated afferents. The threshold for warm sensation reflects the function of warm-specific small unmyelinated afferent channels. Cold pain and heat pain thresholds test the function of unmyelinated C-fiber, polymodal nociceptors, and, to a lesser extent, Aδ-fiber nociceptors. Certain abnormal patterns are characteristic of dysfunction of small-caliber peripheral nerve afferents. To obtain maximal information from a quantitative somatosensory thermotest, it is necessary to test for cold, pain, and heat sensations, which is mandatory in the evaluation of painful syndromes. Quantitative sensory testing performed at different sites along an extremity in patients with polyneuropathy yields useful information about staging of the pathologic process along the extremity.
The quantitative pseudomotor axon reflex test (QSART) is a quantitative thermoregulatory sweat test. It has been used to detect postganglionic pseudomotor failure in neuropathies and preganglionic neuropathies with presumed trans-synaptic degeneration. In patients with distal small-fiber neuropathy, it is the most sensitive diagnostic test. Various commercial devices have been used to differentiate axonal from demyelinating polyneuropathy.
Clinical Correlations of Electromyographic Testing
Clinical correlations can be based on a careful history, clinical examination, and electrodiagnostic studies. Electrodiagnostic studies are best for distinguishing neuropathy from myopathy and determining whether a neuropathy is generalized axonal, demyelinating, mixed, or focal, thus giving important clues about the cause. Furthermore, nerve trauma can be monitored serially to determine recovery. In many instances, a diagnosis of plexopathy or radiculopathy can be made.
Nerve Trauma
Often after an injury, such as a laceration, the nerve is completely severed. At rest, denervation potentials are recorded in the muscles supplied by that nerve in the form of positive sharp waves or fibrillation potentials, and on EMG, no motor unit action potentials are seen. Sometimes, however, an injury is incomplete and the type of nerve lesion is uncertain.
Neurapraxia
Neurapraxia is the mildest form of nerve injury. It consists of loss of conduction without associated changes in axonal structure. This form of conduction block often occurs with compressive or ischemic nerve injuries, such as a mild entrapment syndrome or compression (e.g., “Saturday night palsy”). In neurapraxic injuries, focal demyelination occurs. Serial nerve conduction determinations along the course of the nerve enable one to locate the site of the conduction block. The prognosis for complete recovery is generally good, and healing occurs within days or weeks, barring further injury.
Axonotmesis
In axonotmesis, a more severe form of nerve injury, the axon is disrupted in its myelin sheath. The neural tube, which consists of the endoperineurium and epineurium, remains intact. The nerve undergoes wallerian degeneration, with fragmentation of the axon distal to the site of injury. Motor and sensory paralysis occurs along with associated atrophy of the muscles supplied and loss of reflexes. After about 4 to 5 days, the distal segments of the nerve become inexcitable. In 1 to 2 weeks, positive sharp waves are seen; fibrillations in the involved musculature occur in 2 to 3 weeks. The intact neural tube forms a lattice for the regenerating axon, and the prognosis for recovery is generally good.
Neurotmesis
Neurotmesis is the most severe form of nerve injury and consists of severe disruption or transection of the nerve. Nerve regeneration and recovery are often incomplete, and surgical reanastomosis may be required. Neuromas may form and are commonly associated with pain. Only serial determinations over time can determine the difference between axonotmesis and neurotmesis.
Nontraumatic Neuropathies
In a patient with a nontraumatic neuropathy, segmental demyelination is generally associated with slowing of NCV and temporal dispersion of evoked responses. With axonal degeneration, however, a reduction in the evoked response amplitudes with mild or minimal slowing of NCV is typical. EMG provides early information regarding reinnervation before clinical recovery is evident. The earliest positive evidence of reinnervation is the appearance during voluntary effort of motor unit potentials that are of low amplitude in the beginning but are highly polyphasic (“nascent”) units. They may be present several weeks before clinical evidence of functional recovery is apparent.
Polyneuropathy
EMG and evaluation of nerve conduction are useful in diagnosing polyneuropathy and in determining whether the pathologic process is axonal and demyelinating. A diagnosis of polyneuropathy is made when abnormal nerve conduction and EMG findings are bilateral and symmetrical.
Generalized peripheral neuropathies frequently associated with pain are noted in Box 14.1 . The following electrodiagnostic findings are characteristic of axonal neuropathy:
- 1.
Abnormally low or absent sensory nerve action potentials and compound muscle action potential amplitudes
- 2.
Normal distal latencies
- 3.
Near-normal motor and sensory conduction velocities
Diabetes mellitus
Polyneuropathy associated with insulinoma
Polyneuropathy associated with nutritional deficiency
Alcohol-nutritional deficiency polyneuropathy
Vasculitis-associated neuropathy
Amyloidosis
Cancer and chemotherapeutic-related neuropathies
Sjögren’s syndrome
Pompe’s disease
Toxic substances (arsenic and thallium)
HIV-related distal symmetrical polyneuropathy
Fabry’s disease
Guillain-Barré syndrome (acute inflammatory demyelinating polyneuropathy)
Cryptogenic sensory or sensorimotor neuropathy
Polyneuropathy secondary to neoplasm, including paraneoplastic syndromes such as acute sensory neuropathy or sensorimotor neuropathy associated with carcinoma
HIV, human immunodeficiency virus.
If a disease process affects the large-diameter axons, some slowing of conduction occurs; the velocity is seldom reduced by more than 20% to 30% of normal. However, fibrillations and positive sharp waves are present in muscles innervated by the affected nerves and are generally worse distally. The feet are more involved than the hand muscles, and the leg muscles are involved more than the arm muscles. Motor unit potentials are decreased in number with deficient recruitment and an incomplete interference pattern. Some motor units have increased amplitude and duration.
In contrast, diffuse demyelinating neuropathy is characterized by a reduction in conduction velocity, usually more than 40% of the normal range. Distal latencies are also prolonged. The sensory nerve action potentials and compound muscle action potentials usually exhibit low amplitudes and temporal dispersion. Needle EMG shows no fibrillations or positive sharp waves unless secondary axonal degeneration is present. In a pure demyelinating neuropathy, there is no denervation of muscle fibers. Motor units are decreased in number; the decreased recruitment is attributable to conduction block in some fibers. Usually, no significant change in the duration, amplitude, or morphology of motor units occurs, but the number of polyphasic potentials may be increased if the terminal axons have become demyelinated.
Once it has been determined by EMG whether a neuropathy is primarily axonal or demyelinating, one can then consider clinically which neuropathies are diffusely axonal and which are demyelinating. Subacute and chronic diffuse axonal types include most toxic and nutritional neuropathies, uremia, diabetes, hypothyroidism, HIV infection, Lyme disease, paraneoplastic disease, dysproteinemia, and amyloidosis.
Demyelinating polyneuropathies include hereditary motor and sensory neuropathy types I and III, Refsum’s disease, multifocal leukodystrophy, and Krabbe’s disease. Acute nonuniform demyelinating diseases include Guillain-Barré syndrome, diphtheria, and acute arsenic intoxication, whereas chronic versions include inflammatory demyelinating peripheral neuropathy, idiopathic disease, and neuropathies accompanying HIV disease, as well as various paraproteinemias, dysproteinemias, and osteosclerotic myeloma.
Mononeuropathies and Entrapment Neuropathies
With an entrapment neuropathy, the most commonly involved nerves are the median, ulnar, radial, common peroneal, and tibial. Entities such as trauma, vasculitis, diabetes mellitus, leprosy, and sarcoidosis can affect any nerve in the body. Electrophysiologic studies are of great assistance in localizing the lesion in the individual nerve and in differentiating mononeuropathy from diffuse polyneuropathy, plexopathy, and radiculopathy.
Median Nerve
The median nerve is most commonly entrapped at the wrist as it passes through the carpal tunnel, but it may also be injured at the elbow where it passes between the two heads of the pronator teres or, less frequently, is compressed by a dense band of connective tissue (the ligament of Struthers immediately above the elbow) (see Fig. 14.9 ). The median nerve is derived from the C6 through T1 nerve roots (lateral and medial cords of the brachial plexus). The diagnosis of carpal tunnel syndrome is made by demonstrating localized slowing of sensory and motor conduction across the wrist as evidenced by prolonged sensory and motor distal latencies. In addition, with late changes there may be denervation in the form of fibrillations, positive sharp waves, and reduced motor units with polyphasia in the hand muscles innervated by the median nerve. The need for reference values for special populations, including diabetics and active workers, has been emphasized in a recent AANEM monograph.
A recent review has emphasized that carpal tunnel syndrome may represent a focal intracanal condition such as pregnancy, lipoma, an arterial condition, and the hereditary neuropathy of amyloidosis.
Pronator Teres and Anterior Interosseous Syndromes
The pronator teres and anterior interosseous syndromes consist of proximal compression or entrapment neuropathies of the median nerve. Patients with pronator teres syndrome may also have normal distal latency but no evidence of denervation in the median-innervated hand and forearm muscles except for the pronator teres. The anterior interosseous nerve is a motor branch of the median nerve, with its origin just distal to the pronator teres.
Ulnar Nerve
The ulnar nerve is derived from the C8 and T1 cervical nerve roots (medial cord of the brachial plexus). It is usually injured at the elbow but occasionally at the wrist in the canal of Guyon or deep in the palm (silver beater’s palsy). EMG helps differentiate C8 and T1 radiculopathies from plexopathy or more distal ulnar nerve palsy (see Fig. 14.8 ). When the lesion is in the wrist at the canal of Guyon, usually both sensory and motor fibers are involved and the amplitude of the sensory nerve action potential and muscle action potential is reduced. Distal sensory and motor latency across the wrist is prolonged, and there is no focal slowing of motor NCV or decrement in compound muscle action potential across the elbow. With a lesion in the deep palmar branch, no sensory abnormality occurs and all the changes are in the motor distribution distal to the lesion. When the abnormality is at the elbow, there may be localized slowing of NCV across the elbow, often as much as 25% to 40% below normal. Normal values may depend on the method used (arm straight vs. arm bent). The sensory potential may be affected, as may EMG findings in the ulnar hand muscles.
Radial Nerve
The radial nerve is a continuation of the posterior cord of the brachial plexus and receives fibers from the C5 to C8 cervical roots. It is usually involved at the spiral groove of the humerus, often secondary to a humeral fracture. With a lesion at the spiral groove, the triceps muscle is noted to be spared on EMG, but all the extensor muscles of the forearm are involved. An isolated superficial radial nerve palsy sometimes occurs at the wrist, with the only abnormality being in the radial sensory nerve action potential.
Posterior Interosseous Syndrome
The posterior interosseous nerve syndrome (sometimes known as complicated lateral epicondylitis ) is caused by entrapment of this branch of the radial nerve at the arcade of Fröhse between the two heads of the supinator. EMG shows involvement of the extensor carpi ulnaris, extensor digitorum longus, extensor pollicis longus, and extensor indicis with sparing of the more proximal supinator and extensor carpi radialis longus and brevis. Sensation is unaffected.
Common Peroneal Nerve
The common peroneal nerve is derived from the L4 through S1 roots but primarily from L5. It may be compressed at the head of the fibula. Peroneal NCSs show reduced compound action potentials, as recorded from the extensor digitorum brevis on stimulation above the fibular head, and normal compound action potentials below the fibular head and at the ankle.
Posterior Tibial Nerve at the Ankle
The posterior tibial nerve is derived from the L4 through S3 roots and may be compressed in the tarsal tunnel. NCSs show prolongation of the distal motor and sensory latency of the tibial nerve. There may be EMG changes in the appropriate foot muscles. This syndrome is relatively uncommon.
Sciatic Nerve
The sciatic nerve arises from the L4, L5, S1, S2, and S3 nerve roots. A controversial syndrome is entrapment by the piriformis muscle as it passes through the greater sciatic notch. A lesion of the sciatic nerve is defined and localized by detailed needle examination of muscle in the lower limb.
Other Uncommon Neuropathies
There are numerous potential mononeuropathies, including those involving the long thoracic nerve, dorsal scapular nerve, suprascapular nerve, musculocutaneous nerves, and axillary nerves in the shoulder girdle and upper extremity, as well as those in the pelvic girdle, including the femoral, obturator, saphenous, lateral femoral cutaneous, genitofemoral, ilioinguinal, and superior and inferior gluteal nerves. Needle EMG reveals denervation changes in muscle innervated by individual nerves. NCSs are rarely useful in their evaluation.
Radiculopathies
Radiculopathies are diseases of the nerve roots and must be differentiated from plexopathies, as well as from complex individual nerve root lesions. Roots are commonly involved by compression, especially in the cervical and lumbar region, but they may also be involved in diseases such as diabetes mellitus, herpes zoster, carcinomatous infiltration, and lymphomatous infiltration of nerves, as well as by rare sarcoidosis and infectious processes. Determination of motor and sensory nerve conduction is rarely useful because the lesion in a radiculopathy is proximal to the dorsal root ganglion and motor conduction studies are usually normal, although they may be reduced in amplitude if the lesion is severe enough to cause axonal loss. The H-reflex is absent or latency-delayed when the S1 root is involved. Typically, nerve root lesions are identified by abnormal needle examination results in the appropriate paraspinal and limb muscles. Because most limb muscles are supplied by more than one nerve root ( Tables 14.2 and 14.3 ), a normal study does not exclude the diagnosis of radiculopathy; however, when the findings on EMG are abnormal, they provide objective evidence of functional impairment in the nerve root and localize the lesion to one or more roots in addition to revealing the severity of involvement.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
