FOCAL MONONEUROPATHIES

Compression, traction, laceration, thermal, or chemical injury may damage one or more components of the peripheral nerves, including the myelin, axons, or supporting nerve structures (endoneurium, epineurium, and perineurium). The pathophysiologic responses to peripheral nerve injuries have a limited repertoire; that is demyelination, axon loss, or a combination of both.

Demyelinative Mononeuropathy

With focal injury to myelin, conduction along the affected nerve fiber is altered. This may result in slowing of conduction or conduction block along the nerve fibers or a combination of both.

1. Focal slowing. This is usually the result of widening of the nodes of Ranvier (paranodal demyelination). Focal slowing may be synchronized when demyelination affects all the large myelinated fibers equally. When the focal lesion is distal, there is prolongation of distal and proximal latencies while the proximal conduction velocity remains normal. If the focal lesion is between the distal and proximal stimulation sites, there is prolongation of proximal latency only resulting in slowing in proximal conduction velocity while the distal latency remains normal (Figure 4-1). With lesions manifesting as focal synchronized slowing, the CMAP amplitudes, durations, and areas remain normal and do not change significantly following proximal and distal stimulation. Desynchronized (differential) slowing occurs when conduction time is reduced at the lesion site along a variable number of the medium or small nerve fibers (average or slower conducting axons). Here, the CMAP is dispersed with prolonged duration on stimulations proximal to the lesion (Figure 4-2). The latency and conduction velocity along the injury site remain normal, since at least some of the fastest conducting axons are spared. When the largest axons are also affected, the dispersed CMAP with prolonged duration is also accompanied by slowing of distal latency (in distal lesions) or conduction velocity (in proximal lesions).

2. Conduction block. This is usually the result of focal loss of one or more myelin segment (segmental or internodal demyelination) which leads to interruption of action potential transmission. A nerve lesion manifesting with conduction block is best localized when it can be bracketed by two stimulation points, one distal to the site of injury and one proximal. In conduction block, stimulation distal to the lesion elicits a normal CMAP, whereas proximal stimulation elicits a response with reduced amplitude (partial conduction block) or absent response (complete conduction block) (Figure 4-3). The percentage drop in amplitude and area (amplitude or area decay) are calculated as follows:

Figure 4-1 Nerve conduction studies showing focal slowing in distal segment (a) resulting in slowing of distal latency only, and in proximal segment (b) resulting in slowing of conduction velocity only.

(Reprinted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

Figure 4-2 Nerve conduction studies showing desynchronized slowing. The response with proximal stimulation is dispersed consistent with differential slowing of nerve fibers in the proximal nerve segment.

(Reprinted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

Figure 4-3 Nerve conduction studies showing partial or complete conduction blocks in the proximal segment of the nerve.

(Reprinted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

There are several limitations to the definitive diagnosis of demyelinative conduction block:

(A) A reduced CMAP size may result from phase cancellation between action potentials peaks of opposite polarity because of abnormally increased temporal dispersion. Such excessive desynchronization often develops in acquired demyelinative neuropathies. If the distal and proximal responses have dissimilar waveforms, the discrepancy in amplitude between the two may represent in part a phase cancellation rather than true conduction block. Hence, in true partial conduction block, the significant drop of CMAP amplitude, with stimulation proximal to the lesion site compared with the CMAP distal to it, should not be accompanied by significant (>15%) prolongation of CMAP duration and should be supported by significant drop in CMAP area. In general, more than 50% decrease of the CMAP amplitude across the lesion along with more than 50% drop in CMAP area, are definitive signs of conduction block. A 20–50% drop CMAP amplitude and area may be consistent with conduction block in situations when the two stimulation sites are close (usually 10 cm or less) such as across the fibular neck or elbow (Table 4-1).

(B) A distal demyelinating lesion, causing conduction block of the nerve segment between the most distal possible stimulating point and the recording site, manifest as unelicitable or low CMAP amplitude at both distal and proximal stimulation sites. This finding mimics the findings encountered with axonal degeneration (see axon-loss mononeuropathy). Such a demyelinative lesion could only be distinguished from axon loss lesion by repeating the NCS after several weeks. Often in demyelinative lesions, the distal CMAP improves rapidly, a finding that is not consistent with axon loss.

(C) With proximal demyelinative conduction block lesions, it may not be technically possible to bracket the lesion with two stimulation sites since the lesion is extremely proximal. Examples include demyelinating lesions at the root level, femoral nerve, facial nerve, sciatic nerve, or lumbosacral plexus. In these situations, stimulation distal to the lesion can only be achieved. The diagnosis of conduction block is made only by inference, when a normal or nearly normal CMAP is coupled with significant reduction of recruitment on needle EMG of the recorded muscle.

(D) The prominent temporal dispersion due to phase cancellation seen in evaluating normal SNAPs precludes the use of these potentials in the diagnosis of conduction block (see Chapter 2, Figure 2-15).

(E) Conduction block pattern is often seen with acute axonal loss nerve lesions before the completion of wallerian degeneration (see axon-loss mononeuropathy).

3. Focal slowing with conduction block. Demyelinating lesions presenting with both focal slowing and partial conduction block are not common. They are usually seen at chronic entrapment sites such ulnar nerve lesions across the elbow. In such situations, a significant number of fibers have internodal demyelination resulting in conduction block (drop in amplitude and area across the abnormal segment) while other fibers have paranodal demyelination manifesting as focal slowing of the same nerve segment.

Table 4-1 Electrodiagnosis of Conduction Block

Definite in Any Nerve* 1. ≥50% drop in CMAP amplitude with a 15% prolongation of CMAP duration and with ≥50% drop in CMAP area 2. ≥30% drop in amplitude and area over a short nerve segment (e.g., radial across the spiral groove, ulnar across the elbow, peroneal across the fibular head)

Possible in Median, Ulnar, and Peroneal Nerves Only

1. 20–50% drop in CMAP amplitude with ≤15% prolongation of CMAP duration and with 20–50% drop in CMAP area

* Caution should be taken in evaluating the tibial nerve, where stimulation at the knee can be submaximal, resulting in 50% or at times greater than 50% drop in amplitude and area, especially in overweight and very tall patients.

Axon-Loss Mononeuropathy

Following acute focal axonal damage, the distal nerve segment undergoes wallerian degeneration. However, early after axonal transaction, the distal axon remains excitable. Hence, stimulation distal to the lesion elicits a normal CMAP, whereas proximal stimulation elicits an absent response (complete conduction block) when the lesion is total and reduced CMAP amplitude (partial conduction block) when the lesion is incomplete (Figure 4-4). In an attempt to distinguish this pattern from a demyelinative conduction block, some refer to this pattern as an axonal noncontinuity, early axon loss, or axon discontinuity conduction block.

Figure 4-4 Nerve conduction studies in partial axon loss lesion. Initially, there is an axon discontinuity conduction block. Following wallerian degeneration, the distal response drop in amplitude and the subsequent study showed uniformly low amplitude CMAP irrespective of the site of stimulation.

(Reprinted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

Wallerian degeneration of the axons distal to the nerve lesions is completed in 7–11 days. In the first 1–2 days, the distal CMAP and SNAP are normal. The distal CMAP amplitude then decreases and reaches its nadir in 5–6 days, while the distal SNAP amplitude lags slightly behind. It starts declining in amplitude after 4–5 days and reaches its nadir in 10–11 days (Figure 4-5). The earlier decline of the CMAP amplitude comparing to the SNAP amplitude following axon-loss nerve lesion is likely related to the early neuromuscular transmission failure that affects the recording of the CMAP amplitudes only. This is supported by the fact that MNAPs, recorded directly from nerve trunks, follow the time course of SNAPs.

Figure 4-5 Effect of wallerian degeneration on distal CMAP and SNAP amplitudes following acute nerve lesion.

On motor NCS, a conduction block is present soon after axonal injury. However, as the distal axons undergo wallerian degeneration, this is replaced by unelicitable or low CMAP amplitudes with both distal and proximal stimulations corresponding to complete or partial motor axonal loss lesions respectively (see Figure 4-4). At this time, the distal CMAP amplitude is a reliable semiquantitative estimate of the amount of axonal loss in peripheral nerve lesions. In the chronic phases of partial axonal nerve lesions with reinnervation via collateral sprouting, the CMAP may improve to reach normal or near normal values giving a false indication of a milder degree of original axonal loss.

When the electrodiagnostic study is done early after an acute peripheral nerve lesion, it should be repeated at 10–11 days or later (or 5 days or later in purely motor nerves) in order to distinguish between conduction block caused by demyelination versus axonal loss, and to assess the extent of axon loss if present. Following this period of wallerian degeneration, stimulating the nerve below the lesion results in absent or reduced CMAP amplitude since degenerating axons would have lost their excitability. An absent or reduced CMAP amplitude from stimulation above or below the lesion indicates complete or partial axonal loss respectively. In demyelinating lesions, the distal CMAP remains unchanged with persistent conduction block across the lesion. In mixed lesions, the distal CMAP drops but remains significantly higher than the proximal implying both axon loss and segmental demyelination.

In partial axon-loss peripheral nerve lesions, the distal latencies and conduction velocities remain normal or are borderline. Selective loss of fast-conducting fibers associated with more than a 50% reduction in mean CMAP amplitude can slow conduction velocity to 80% of normal value because the velocity represents the remaining slow-conducting fibers. Motor conduction velocity may be occasionally slowed to 70% of normal value, when there is severe axonal loss with marked reduction of CMAP amplitude to less than 10% of normal.

Needle EMG is useful in assessing the progress of reinnervation of axon loss peripheral nerve lesions that may occur spontaneously or after nerve repair. Although collateral sprouting in partial axon loss lesions starts as early as 1–2 days after a nerve lesion, the early signs of reinnervation may first become evident on needle EMG one month later, but are usually definite by 2–3 months postinjury. MUAP morphology helps assessing the process of muscle fiber reinnervation that occurs following collateral sprouting and proximodistal regeneration of nerve fibers from the site of the injury. Collateral sprouting causes first an increased number of MUAP turns and phases followed by an increased duration and amplitude of MUAPs, while early proximodistal regeneration of nerve fibers in severe axon loss lesions often manifests by recording brief, small, and highly polyphasic (nascent) MUAPs. MUAPs tend to become longer in duration and higher in amplitude with the passage of time due to improved synchrony of muscle fiber action potentials.

In contrast to demyelinating or mixed mononeuropathies, pure axon-loss peripheral nerve lesions cannot be localized by NCSs when studied after the completion of wallerian degeneration, since they are not associated with focal conduction slowing or block. The identification of conduction block in the early days of axonal loss is extremely helpful in localizing a peripheral nerve injury. Waiting for the completion of wallerian degeneration results in diffusely low or unevoked CMAPs (regardless of stimulation site), which does not allow for accurate localization of the injury site. Localizing a purely axon-loss mononeuropathy after the completion of wallerian degeneration depends on needle EMG, with principles that are similar to manual muscle strength testing used during the neurological examination. Typically, the needle EMG reveals neurogenic changes (fibrillation potentials, reduced MUAP recruitment, chronic neurogenic MUAP morphology changes) that are limited to muscles innervated by the injured nerve distal to the site of the lesion (Figure 4-6). In contrast, muscles innervated proximal to the lesion remain normal. Unfortunately, attempting to localize axon loss lesions solely by needle EMG has several shortcomings that may result in poor localization or, sometimes, mislocalization of the site of the nerve lesion. These include the following scenarios:

Figure 4-6 Localization of peripheral nerve lesion using needle EMG. Muscles distal to the lesion reveals abnormal neurogenic findings (+ +) while proximal muscles are normal (O).

(Adapted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20: 305–338, with permission.)

Figure 4-7 Localization of peripheral nerve lesion using needle EMG. The diagram on the left shows an “ideal” nerve with branches separated by short nerve segments (e.g., the radial nerve) allowing precise localization. In contrast, the diagram on the right depicts a nerve with a long segment that gives no motor branches (e.g., the median nerve in the arm) resulting in a suboptimal localization of lesion to a long nerve segment (+ + are denervated muscles while O are normal muscles).

(Adapted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20: 305–338, with permission.)

Figure 4-8 Fascicular peripheral nerve lesion leading to mislocalization of lesion more distally (+ + are denervated muscles while O are normal muscles).

(Adapted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

Figure 4-9 Chronic nerve lesion. The diagram on the left shows that muscles located distal to the lesion and tested in the acute and subacute phase of nerve injury showed signs of denervation. In contrast, the diagram on the right reveals with the passage of time (months to years) effective reinnervation and remodeling of several muscles close to the lesion site leading to mislocalization of the lesion more distally (+ and ++ are denervated muscles while O are normal muscles).

(Adapted from Wilbourn AJ. Nerve conduction studies. Types, components, abnormalities and value in localization. Neurol Clin 2002;20:305–338, with permission.)

RADICULOPATHIES AND PLEXOPATHIES

Radiculopathies are, by definition, lesions of the ventral or dorsal roots or both occurring within the spinal canal space. In contrast, plexopathies are lesions that involve the peripheral nerve extraspinally. Since the dorsal root ganglia are usually located outside of the spinal canal and within the intervertebral foramina, radiculopathies are considered preganglionic lesions while plexopathies are postganglionic.

The dorsal root ganglia contain unipolar sensory neurons with a peripheral and a central axon. In radiculopathies associated with axonal loss due to lesions of the proximal sensory axons, the distal sensory axons do not degenerate since the dorsal root neurons usually escape injury. Hence, the SNAP remains normal despite sensory loss and degeneration of proximal sensory axons. When the motor axons within the ventral roots are also injured, radiculopathies exhibits signs of motor axon degeneration including abnormal needle EMG and, when severe, low-amplitude CMAPs.

Needle EMG is the most sensitive and specific electrodiagnostic test for the identification of cervical and lumbosacral radiculopathies, particularly those associated with axon loss. Needle EMG is also useful in the accurate localization of the level of the root lesion. Finding signs of denervation and reinnervation (fibrillation potentials, decrease recruitment, and long-duration, high-amplitude polyphasic MUAPs) in a segmental myotomal distribution (i.e., in muscles innervated by the same roots via more than one peripheral nerve), with or without denervation of the paraspinal muscles localize the lower motor neuron lesion to the root level. A normal SNAP of the corresponding dermatome ensures that the lesion is within the spinal canal (i.e., proximal to the dorsal root ganglia). For example, in a C7 radiculopathy, the triceps (radial nerve) and pronator teres (median nerve) are often abnormal on needle EMG, with or without the cervical paraspinal muscles, and the median SNAP recording middle finger is normal.

In contrast to intraspinal canal root lesions, axon-loss extraspinal plexopathies affect the CMAP as well as the SNAP amplitudes when mixed nerves undergo wallerian degeneration. Abnormal SNAPs are not compatible with root lesions (preganglionic), but consistent with lesions affecting the brachial plexus (postganglionic). These findings are particularly important in brachial plexus traction injuries that may mimic root avulsions. In avulsions, the dorsal root ganglia remain intact despite severe sensory loss and the peripheral sensory axons do not undergo wallerian degeneration. Hence, the SNAPs in root avulsions are spared while they are low in amplitude or absent in brachial plexopathies.

GENERALIZED POLYNEUROPATHIES

Nerve conduction studies are essential in the diagnosis of peripheral polyneuropathies, and in establishing the type of fiber(s) affected (large fiber sensory, motor, or both). Most importantly, NCSs often can identify the primary pathological process of peripheral polyneuropathy (axonal loss or segmental demyelination), an important step in establishing the etiological diagnosis of the various peripheral polyneuropathies.

Demyelinating Polyneuropathies

The electrophysiologic hallmark of these polyneuropathies is a widespread increase in conduction time due to impaired saltatory conduction. Hence, the NCSs are characterized by significant slowing of conduction velocities (<75% of lower limit of normal) and distal latencies (>130% of upper limit of normal).

With distal stimulation, the CMAP amplitude is mildly or moderately reduced because of abnormal temporal dispersion and phase cancellation, and the distal latency is delayed because of demyelination. With more proximal stimulation, the CMAP amplitude is lower due to temporal dispersion and conduction block along some fibers. The proximal conduction velocity is markedly slowed because of increased probability for the nerve action potentials to pass through demyelinated segments (Figure 4-10C).

Figure 4-10 Computerized model of peripheral motor nerve in normal (A), axonal degeneration (B), and segmental demyelination (C).

(Reprinted from Brown WF, Bolton CF, eds. Clinical electromyography. Boston, MA: Butterworth-Heinemann, 1989, with permission.)

Chronic demyelinating polyneuropathies may be further distinguished by NCS into inherited and acquired polyneuropathies. Inherited demyelinating polyneuropathies such as Charcot-Marie-Tooth disease type I, are characterized by uniform slowing along various segments of individual nerves and adjoining nerves. The abnormalities are usually symmetrical without accompanying conduction blocks (except possibly at compressive sites). In contrast, acquired demyelinating polyneuropathies, such as chronic inflammatory demyelinating polyneuropathy, often have asymmetric nerve conductions, even when there is no apparent clinical asymmetry. In addition, multifocal conduction blocks and excessive temporal dispersions at nonentrapment sites are characteristics for acquired demyelinating polyneuropathies.

In demyelinating polyneuropathies, the needle EMG may show signs of mild axonal loss manifested by fibrillation potentials and reinnervated MUAPs.

ANTERIOR HORN CELL DISORDERS

There are three reasons for performing electrodiagnostic studies in patients with suspected amyotrophic lateral sclerosis: (1) to confirm lower motor neuron dysfunction in clinically affected regions; (2) to detect electrophysiologic evidence of lower motor neuron dysfunction in clinically uninvolved regions; and (3) to exclude other pathophysiologic processes. A disadvantage of the clinical EMG study is that it can only evaluate lower motor neuron degeneration while upper motor neuron degeneration can only be assessed clinically. Hence, the diagnosis of amyotrophic lateral sclerosis with evidence of upper and lower motor neuron degeneration is often based on the clinical evaluation with the electrodiagnostic study playing only a supporting role.

In patients with suspected motor neuron disease, sensory NCSs are usually normal. Motor NCSs are either normal or yield low CMAP amplitudes consistent with motor neuronal loss. There are no motor conduction blocks and the motor conduction velocities are normal or slightly slowed not below 70% of the lower limits of normal. In patients with suspected motor neuron disease, NCSs are most useful in excluding other neuromuscular diagnosis such as polyneuropathies, multifocal motor neuropathy, or neuromuscular junction disorders.

Needle EMG is the most important electrodiagnostic study for providing evidence of generalized lower motor neuron degeneration. Early in the course of the illness, denervation in clinically normal muscles and limbs is most useful in establishing early dissemination of denervation. Needle EMG in amyotrophic lateral sclerosis often shows signs of active denervation (fibrillation and fasciculation potentials), chronic denervation (reinnervated and unstable MUAPs), and reduced MUAP recruitment. Lambert’s original criteria for diagnosis include detecting fibrillation and fasciculation potentials in muscles of the lower as well as the upper extremities or in the extremities as well as the head. These criteria evolved over the years into denervation at least three extremities or two extremities and cranial muscles (the head and neck considered an extremity). Although lower motor neuron degeneration ultimately affects almost the entire neuraxis (brainstem and cervical, thoracic, or lumbosacral segments of spinal cord), the early phases of the illness are often characterized by limited and more focal weakness. The revised El Escorial criteria recommend that needle EMG signs of lower motor neuron degeneration should be present in at least two of the four central nervous system regions, i.e., the brainstem, cervical, thoracic, or lumbosacral regions.

MYOPATHIES

Insertional activity is usually normal or increased except in the late stage of the disease when it is reduced by atrophy and fibrosis. Spontaneous activity is absent except in necrotizing myopathies (such as inflammatory myopathies and muscular dystrophies). MUAP amplitude and duration are reduced because of random loss of fibers from the motor unit. Split muscle fibers and regeneration of muscle fibers sometimes accounts for satellite potentials and polyphasia. Early recruitment is common because more motor units are needed to maintain a given force in compensation for the small size of individual units.

A disadvantage of the electrodiagnosis of myopathies is that the EMG findings in myopathy are not always specific to make a final diagnosis. Exceptions include conditions that are associated with (1) myotonia such as myotonic dystrophies, myotonia congenita, paramyotonia congenita, hyperkalemic periodic paralysis, acid maltase deficiency, and some toxic myopathies (such as colchicine), or (2) fibrillation potentials which occur in inflammatory myopathies, critical illness myopathy, and progressive muscular dystrophies. Another disadvantage of the needle EMG is that it is either normal or has only subtle abnormalities in many myopathies particularly those not usually associated with myonecrosis, such as the metabolic and endocrine myopathies. Hence, a normal needle EMG does not exclude a myopathy.

In polymyositis and dermatomyositis, it is essential to recognize the changing pattern on needle EMG at diagnosis, following treatment, and during relapse. Fibrillation potentials appear first at diagnosis or relapse and disappear early during remission. Abnormal MUAP morphology becomes evident later and lasts longer to resolve. The presence of fibrillation potentials is also helpful in differentiating exacerbation of myositis from a corticosteroid-induced myopathy.

NEUROMUSCULAR JUNCTION DISORDERS

In myasthenia gravis as well as other postsynaptic neuromuscular junction disorders, the baseline (resting) CMAP is normal. Slow repetitive nerve stimulation (RNS) results a CMAP decrement that often improves or corrects after brief exercise. The diagnostic sensitivity of RNS is much higher recording proximal than distal muscles. The decrement often improves or is repaired after exercise (“post-tetanic facilitation”), and worsens several minutes after exercise (“postexercise exhaustion”). If done, rapid RNS (or postexrecise CMAP) causes no change of CMAP or, in severe situations, a CMAP decrement.

In Lambert-Eaton myasthenic syndrome, botulism, and other presynaptic disorders, the baseline (resting) CMAP amplitude is low. With slow RNS, there is further CMAP decrement. However, with rapid RNS or postexercise CMAP, there is a CMAP increment, which is typically higher than 200% in Lambert-Eaton myasthenic syndrome, and is usually 30–100% in patients with botulism.

UPPER MOTOR NEURON LESIONS

Patients with upper motor neuron lesions have normal NCSs and needle EMG including normal insertional activity, no spontaneous activity at rest, and normal MUAP morphology. The only abnormality is a reduced interference pattern with poor activation of MUAPs (slow rate of motor unit discharge). Recruitment, measured by either recruitment frequency or ratio, is normal. Hysterical weakness or poor effort produces a similar pattern, except that motor unit firing may be irregular.