- Email: [email protected]
Repetitive nerve stimulation and exercise testing
Phys Med Rehabil Clin N Am 14 (2003) 185–206
Repetitive nerve stimulation and exercise testing Barbara E. Shapiro, MD, PhD*, David C. Preston, MD Department of Neurology, University Hospitals of Cleveland, Case Western Reserve University, 11100 Euclid Avenue, Cleveland, OH 44106-5098, USA
Repetitive nerve stimulation (RNS) and exercise testing have a key role in the electrodiagnostic evaluation of patients with suspected neuromuscular junction (NMJ) disorders, especially myasthenia gravis, Lambert-Eaton myasthenic syndrome, and botulism [1–5]. In addition, these studies are often useful in patients with rare disorders of skeletal muscle membrane excitability, including paramyotonia congenita, myotonia congenita, myotonic dystrophy, and the periodic paralyses [6,7]. The technique dates back to the late 1800s when Jolly first made visual observations of muscle movement following RNS. Unlike contemporary methods, his initial studies were performed with submaximal stimuli, using mechanical rather than electrical measurements. Using these techniques, Jolly noted a decremental response following RNS in patients with myasthenia gravis and correctly concluded that the disorder had a peripheral rather than central origin. Since that time, RNS and exercise testing have been advanced and validated to become one of the most useful tests in the evaluation of patients with disorders of the NMJ and muscle membrane excitability [8,9]. In any patient presenting with symptoms of a possible NMJ disorder (eg, fatigability, proximal weakness, dysphagia, dysarthria, or ocular abnormalities), one should perform RNS as part of the electrodiagnostic evaluation. The pattern of responses to slow (2–3 Hz) and rapid RNS (30–50 Hz) is useful in determining whether a NMJ disorder is present and whether it is presynaptic or postsynaptic in origin [5,8]. In addition to RNS, exercise testing has an important role in the evaluation of patients with suspected NMJ disorders. Abnormalities on slow RNS are often more pronounced several minutes following exercise testing and can increase the sensitivity of the study. * Corresponding author. E-mail address: [email protected] (B.E. Shapiro). 1047-9651/03/$ – see front matter Ó 2003, Elsevier Inc. All rights reserved. doi:10.1016/S1047-9651(02)00129-8
186
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Brief maximal voluntary exercise can be used to demonstrate many of the same effects as rapid RNS, which are often key in detecting a presynaptic NMJ disorder. Rapid RNS is painful and difficult to tolerate, whereas exercise testing has the distinct advantage of being painless and is preferable to rapid RNS in patients who are alert and cooperative [10]. When a cooperative subject is asked to contract a muscle voluntarily at maximum force, motor units fire at their maximal firing frequency, typically 30 to 50 Hz. Brief maximal voluntary exercise can be used as a painless substitute to demonstrate the same effects as rapid RNS. Less commonly, exercise testing is employed in the electrodiagnostic evaluation of the rare patient with a suspected disorder of muscle membrane excitability. Exercise testing can be performed over a short (ie, 10 seconds) or prolonged (ie, several minutes) period. In some of these disorders, muscle membrane inexcitability may follow exercise testing, with a corresponding and progressive decline in the compound muscle action potential (CMAP) amplitude as weakness develops [7,11,12]. Exercise testing is useful not only in detecting a muscle membrane excitability disorder but can often aid in further narrowing the differential diagnosis based on the pattern of abnormalities [13]. In the electromyography (EMG) laboratory, the effects of RNS and exercise testing on the CMAP are studied. Analysis of any decremental or incremental response forms the basis of the study. To interpret these responses accurately, one must understand normal NMJ physiology and the effects of RNS and exercise on a single NMJ and its associated muscle fiber. This knowledge can then be used in the EMG laboratory to predict accurately the effect of RNS and exercise testing on the CMAP in normal subjects and in patients with NMJ and skeletal muscle membrane excitability disorders.
Normal neuromuscular junction physiology The NMJ forms an electrical-chemical-electrical link between nerve and muscle [14] (Fig. 1). The chemical neurotransmitter at the NMJ is acetylcholine (ACh). ACh molecules are packaged as vesicles in discrete units known as quanta within the presynaptic terminal. Each quantum contains approximately 10,000 molecules of ACh, located in three separate stores. The primary or immediately available store consists of approximately 1000 quanta located just beneath the presynaptic nerve terminal membrane. As its name implies, this store is immediately available for release. The secondary or mobilization store consists of approximately 10,000 quanta that can resupply the primary store after a few seconds. A tertiary or reserve store of more than 100,000 quanta is located far from the NMJ in the axon and cell body. When a nerve action potential invades and depolarizes the presynaptic junction, voltage-dependent calcium channels are activated, allowing an
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
187
Fig. 1. Normal NMJ anatomy. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
influx of calcium. This infusion of calcium results in the release of ACh from the presynaptic terminal. The greater the calcium concentration inside the presynaptic terminal, the more quanta are released. ACh then diffuses across the synaptic cleft and binds to ACh receptors on the postsynaptic muscle membrane. The postsynaptic membrane is composed of numerous junctional folds, effectively increasing the surface area of the membrane, with ACh receptors clustered on the crests of the folds [15]. The binding of ACh to its receptors opens ion channels, resulting in a local depolarization, the end-plate potential (EPP) [16]. The size of the EPP is proportional to the amount of ACh that binds to the receptors. If the EPP depolarizes the muscle membrane above threshold, an allor-none muscle fiber action potential is generated and propagated through the muscle fiber, similar to the generation of an all-or-none nerve action potential. Under normal circumstances, the EPP always rises above threshold, resulting in a muscle fiber action potential. The amplitude of the EPP above the threshold value needed to generate a muscle fiber action potential is called the safety factor [17]. ACh is then broken down in the synaptic cleft by the enzyme
188
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
acetylcholinesterase, with the choline subsequently taken back up into the presynaptic terminal to be repackaged into ACh. During slow RNS (2–3 Hz), ACh quanta are progressively depleted from the primary store, and subsequently fewer quanta are released with each successive stimulation. The corresponding EPP decreases in amplitude. In normal subjects, because of the normal safety factor, the EPP remains above threshold, ensuring the generation of a muscle fiber action potential with each stimulation. After the first few seconds, the secondary store begins to replace the depleted quanta with a subsequent rise in the EPP. The physiology of rapid RNS (30–50 Hz) in normal subjects is more complicated. Depletion of quanta from the presynaptic terminal is counterbalanced not only by the mobilization of quanta from the secondary store but also by the accumulation of calcium. Normally, approximately 100 ms are required for calcium to diffuse back out of the presynaptic terminal. If RNS is rapid enough so that new calcium infuses in before the previously infused calcium has diffused back out, calcium continues to accumulate in the presynaptic terminal, resulting in an increased release of quanta. This increased number of quanta results in a correspondingly higher EPP. Nevertheless, in normal subjects, the result is the same as with any other EPP above threshold, that is, an all-or-none muscle fiber action potential. Although the effects of slow and rapid RNS are very different at the molecular level, in normal subjects, the result is the same—the consistent generation of a muscle fiber action potential. In pathologic conditions in which the safety factor is reduced (ie, the baseline EPP is reduced but still above threshold), slow RNS will cause depletion of quanta and may drop the EPP below threshold, resulting in the absence of a muscle fiber action potential. In pathologic conditions in which the baseline EPP is below threshold and a muscle fiber action potential is not generated, rapid RNS may increase the number of quanta released, resulting in a larger EPP such that threshold is reached. Once threshold is reached, a muscle fiber action potential is generated where one had not been present previously. These concepts form the basis of decrements to slow RNS and increments to rapid RNS and brief exercise seen in patients with NMJ disorders.
Repetitive nerve stimulation modeled at the level of the neuromuscular junction and its associated muscle fiber Repetitive nerve stimulation in normal subjects and patients with NMJ disorders can be modeled effectively by making the following three assumptions: m ¼ pn, where m is the number of quanta released during each stimulation; p is the probability of release (effectively proportional to the concentration of calcium), typically approximately 0.2 in normals; and n is the number of quanta in the primary store (at baseline, typically approximately 1000 in normals).
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
189
The secondary store starts to replenish the primary store after 1 to 2 seconds. Calcium requires approximately 100 ms to be pumped out of the presynaptic terminal. If stimulation occurs again before 100 ms (ie, stimulation rate >10 Hz), the calcium concentration increases, the probability of release of ACh quanta increases, and more quanta are released. Slow repetitive nerve stimulation In normal subjects during slow RNS, approximately 20% of the quanta are released from the primary store in the first stimulation and with each subsequent stimulation. If the resultant EPP is above threshold, a muscle fiber action potential is generated. Between the first and fourth stimulation, there is a normal depletion of the primary store, a subsequent decline in the number of quanta released, and a corresponding decrease in the EPP. By the fifth stimulation, sufficient time has elapsed for the secondary store to begin to resupply the primary store. As the number of quanta in the primary store increases at this point, with a corresponding increase in the number of ACh quanta released, a higher EPP is generated. In normal subjects, because the EPP stays above threshold at all times, the generation of a muscle fiber action potential is consistent (Fig. 2A). In contrast, in postsynaptic NMJ disorders, the same number of quanta generates a smaller EPP. Accordingly, the safety factor is reduced. In myasthenia gravis, the most common of the postsynaptic disorders, there are fewer ACh receptors and, accordingly, less binding of ACh [18,19]. The reduced safety factor in conjunction with the normal depletion of quanta results in subsequent EPPs falling below threshold and their corresponding muscle fiber action potentials not being generated (Fig. 2B). In Lambert-Eaton myasthenic syndrome, a presynaptic disorder, the number of quanta in the primary store is normal, and the EPP is normal for the number of quanta released. The abnormalities include the reduced number of ACh quanta released and the low baseline EPP [20,21]. The calcium concentration in the presynaptic terminal is reduced owing to an antibody attack on the voltage-gated calcium channels. The probability of release dramatically falls, with a decrease in the number of quanta released. There is still depletion, although not as marked as in normal subjects or patients with postsynaptic disorders, simply because so few quanta are released that the subsequent amount of depletion cannot be as great. Because the EPP is below threshold at baseline, a muscle fiber action potential is never generated (Fig. 2C). In some presynaptic disorders, the baseline EPP may be low but still above threshold, resulting in a reduced safety factor. In this situation, a muscle fiber action potential may be generated initially but then fails to be generated as the EPP falls below threshold with slow RNS.
190
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 2. End-plate potentials. Dotted line (- - - -) marks the threshold. Shaded EPPs rise above threshold and generate a muscle fiber action potential. (A) 3-Hz RNS, normal NMJ. Note that all of the potentials remain well above threshold despite the normal decline in EPP amplitude (safety factor). (B) 3-Hz RNS, postsynaptic NMJ disorder. Note the lower EPP amplitudes. With further ACh depletion, the last three potentials fall below threshold, and a muscle fiber action potential is not generated. (C ) 3-Hz RNS, presynaptic NMJ disorder. Note that all of the EPPs are below threshold, and no muscle fiber action potentials are generated. The EPP declines in amplitude but not as markedly as in normal subjects or patients with postsynaptic NMJ disorders. (D) 50-Hz RNS, presynaptic NMJ disorder. Note the progressive increment in the EPP amplitude to above threshold and the subsequent generation of muscle fiber action potentials. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
Rapid repetitive nerve stimulation The effects of rapid RNS can be deduced from the three basic assumptions described previously. With rapid RNS, the depletion of quanta is counterbalanced by (1) increased mobilization of quanta from the secondary to the primary store and (2) calcium accumulation in the presynaptic terminal,
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
191
which increases the probability of release. The sum of these influences usually results in a greater number of quanta released and higher EPPs with RNS. In normal subjects, rapid RNS always results in the generation of a muscle fiber action potential, the same as with any EPP above threshold. In patients with postsynaptic NMJ disorders, the EPP will also increase, but, because the EPP is usually above threshold at baseline, the result will still be the generation of a muscle fiber action potential. If the EPP has been lowered, such as after slow RNS, the decreased EPP may be repaired or improved with rapid RNS. If the EPP has dropped below threshold, subsequent rapid RNS may increase the EPP back to above threshold. In patients with presynaptic NMJ disorders, the situation is distinctly different. Because the EPP is abnormally low at baseline and often below threshold, rapid RNS may increase the EPP above threshold so that a muscle fiber action potential is generated where one had not been present previously (Fig. 2D). Repetitive nerve stimulation and brief exercise testing in the electromyography laboratory The previous discussion has pertained to end-plate and individual muscle fiber action potentials. During RNS in the EMG laboratory, all measurements are made on the CMAP, the sum of the individual muscle fiber action potentials generated in a muscle. One infers that the CMAP amplitude and area are proportional to the number of muscle fibers activated. In normal subjects, although the EPP is affected by slow and rapid RNS, the potential always remains above threshold, resulting in the consistent generation of muscle fiber action potentials. In normal subjects, CMAPs generated following either slow or rapid RNS do not change significantly in amplitude or area. Slow repetitive nerve stimulation In patients with NMJ disorders, if the normal EPP safety factor is reduced, slow RNS will result in the depletion of quanta and reduce the amplitude of the EPP. If the EPP of some muscle fibers falls below threshold, those muscle fiber action potentials will not be generated, and the number of individual muscle fiber action potentials will decline. As the number of individual muscle fiber action potentials declines, a decrement of the CMAP amplitude and area likewise occurs. This effect provides the basis for a decremental CMAP response to slow RNS in the EMG laboratory, which reflects fewer EPPs reaching threshold and fewer individual muscle fiber action potentials contributing to the CMAP. Rapid repetitive nerve stimulation In patients with NMJ disorders in whom some EPPs are below threshold at baseline (usually the presynaptic NMJ disorders), rapid RNS can be used to
192
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
facilitate the EPP. If subthreshold EPPs can be brought above threshold, muscle fiber action potentials will be generated where they had not been present previously, and the number of individual muscle fiber action potentials will increase. This effect provides the basis for an incremental CMAP response to rapid RNS seen in the EMG laboratory. As the number of individual muscle fiber action potentials increases, an increment of the CMAP amplitude and area occurs, which reflects more EPPs reaching threshold and more individual muscle fiber action potentials contributing to the CMAP. Incremental responses greater than 100% above baseline in response to rapid RNS are not unusual in presynaptic NMJ disorders. Brief exercise testing In postsynaptic NMJ disorders, brief exercise, just like rapid RNS, results in higher EPPs; however, because the EPP is usually above threshold at baseline, the result is the same, that is, the generation of a muscle fiber action potential. Brief exercise may repair or improve a low EPP that has developed following slow RNS. If the EPP has dropped below threshold, subsequent exercise may increase the EPP back to above threshold. In presynaptic NMJ disorders, brief exercise, like rapid RNS, can often facilitate low EPPs. If the baseline EPP is below threshold, brief exercise may increase the EPP above threshold so that a muscle fiber action potential is generated where one had not been present previously. As is true with rapid RNS, as the number of individual muscle fiber action potentials increases, brief exercise results in an increment of the CMAP amplitude and area. This increment reflects more EPPs reaching threshold and more individual muscle fiber action potentials contributing to the CMAP. Post exercise facilitation and exhaustion The effects of rapid RNS or voluntary exercise occur with brief periods of exercise or rapid RNS, typically 10 seconds. This process is known as postexercise (or posttetanic) facilitation. A phenomenon of postexercise (or posttetanic) exhaustion, which is not as well understood, occurs after a more extended period of exercise. Immediately after an extended period of exercise or rapid RNS (usually 1 minute), EPPs typically increase initially, followed by a subsequent decline in EPPs, continuing and worsening over the next several minutes. In normal subjects, the EPP never falls below threshold because of the normal safety factor. In contrast, in patients with impaired NMJ transmission, when slow RNS is performed 2 to 4 minutes after extended exercise, decrements are more pronounced when compared with the baseline, because many more EPPs no longer reach threshold, and their corresponding muscle fiber action potentials are not generated. The effects of postexercise facilitation and postexercise exhaustion can be demonstrated on the CMAP in patients with NMJ disorders (Fig. 3). Post-
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
193
exercise facilitation can be demonstrated in two situations. Brief exercise can repair EPPs that have been lowered by slow RNS. If the EPPs are facilitated above threshold, muscle fiber action potentials will be generated that were not present previously. Accordingly, a decrement of the CMAP amplitude and area that developed during slow RNS may be lessened or ‘‘repaired’’ (Fig. 3A, B). In presynaptic NMJ disorders such as Lambert-Eaton myasthenic syndrome associated with reduced release of quanta and subthreshold EPPs at baseline, brief exercise can facilitate EPPs above threshold, giving rise to muscle fiber action potentials that were not present previously [22,23]. Accordingly, an increment of the CMAP amplitude and area occurs (Fig. 4). To demonstrate the phenomenon of postexercise exhaustion, the muscle is maximally exercised for 1 minute. Slow RNS is then performed immediately after exercise and at 1, 2, 3, and 4 minutes later. In normal subjects with a normal safety factor, the EPP never falls below threshold, and the CMAP amplitude and area remain stable. In contrast, in patients with impaired NMJ transmission, the decrement in CMAP amplitude and area in response to slow RNS becomes more marked 2 to 4 minutes after a minute of exercise (Fig. 3C, D, E). In this situation, 10 seconds of maximal voluntary exercise will repair the decrement toward normal (postexercise facilitation) (Fig. 3F). An important additional concept known as pseudofacilitation should be appreciated by every electromyographer in regards to brief exercise testing or rapid RNS. In normal subjects, brief maximal exercise often leads to a slight increase in CMAP amplitude, which is essential not to misinterpret as an abnormality. Following brief exercise, EPPs are facilitated; however, because they are above threshold at baseline, the same number of muscle fiber action potentials is generated. Although there is no increase in the actual number of muscle fiber action potentials that summate to create the CMAP, brief maximal exercise causes the muscle fibers to fire more synchronously. Because the CMAP is composed of many summated individual muscle fibers, increased synchrony of firing results in slightly better alignment of the negative and positive phases of the individual muscle fiber action potentials, less normal phase cancellation, and a slightly higher CMAP. This pseudofacilitation results in a small increase in CMAP amplitude but usually with a small decrease in CMAP duration and little change in the CMAP area (Fig. 5). In general, postexercise increments of CMAP amplitude from pseudofacilitation do not exceed 40% of baseline in normal subjects.
Technical factors in repetitive nerve stimulation and exercise testing Close attention to technical factors is critical when performing RNS and exercise testing. Technical factors may result in spurious decrements or increments and the mistaken impression of a NMJ disorder if not carefully controlled.
194
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
195
Fig. 4. Exercise testing in Lambert-Eaton myasthenic syndrome. Stimulating the median nerve supramaximally at the wrist, recording from the abductor pollicis brevis muscle. Top trace shows baseline. Bottom trace shows response immediately after 10 seconds of maximal voluntary exercise. Note the marked increase in CMAP amplitude (postexercise facilitation). Preexercise and postexercise testing, looking for an increment, is always better tolerated by patients than is 50-Hz RNS. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
Immobilization: isometric electrode position is essential The greatest technical difficulty with RNS and exercise testing is the need to immobilize the recording electrode properly over the muscle. If the position of the recording electrode moves in relationship to the muscle during stimulation, the CMAP configuration may change. The goal is to minimize any movement of the limb, stimulator, or recording electrodes during RNS and
b Fig. 3. Postexercise facilitation and exhaustion. 3-Hz RNS in a patient with myasthenia gravis. (A) Decrement of CMAP amplitude at rest. (B) Postexercise facilitation. Decrement of CMAP occurs immediately following 10 seconds of maximal voluntary exercise and subsequently repairs toward normal. (C, D, E) Postexercise exhaustion. Decrements of CMAP at 1, 2, and 3 minutes after 1 minute of maximal voluntary exercise. The decrement becomes progressively more marked over the baseline decrement. (F ) Postexercise facilitation after a decrement. Immediately following another 10 seconds of maximal voluntary exercise, the decrement, which has worsened as a result of postexercise exhaustion, repairs toward normal. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
196
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 5. Pseudofacilitation. When performing RNS following exercise testing, pseudofacilitation is often encountered. Pseudofacilitation is a normal phenomena caused by more synchronous firing of muscle fiber action potentials immediately following brief intense exercise. In the normal subject, 3-Hz RNS results in a 0% CMAP decrement at rest (top trace). Immediately following 10 seconds of maximal voluntary exercise, 3-Hz RNS is repeated (bottom trace). A similar 0% decrement is found; however, the CMAP amplitudes are higher, the durations are shorter, and the areas are similar from the normal effects of pseudofacilitation. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: ButterworthHeinemann; 1998; with permission.)
exercise testing. The recording electrodes should be well secured with tape. If possible, the stimulator should be secured with tape or a Velcro strap and the entire limb secured to a pad or board (Fig. 6). Immobilization is more easily accomplished when stimulating distal nerves such as the median or ulnar. Securing the stimulator and limb is more problematic when testing proximal nerves. Stimuli must be supramaximal The use of submaximal stimulation can create a variety of problems, including spurious CMAP decrements and increments (Fig. 7). A supramaximal stimulus must always be obtained before beginning RNS and exercise testing. Temperature control is essential Repetitive nerve stimulation and exercise testing in the EMG laboratory should be performed with the limb at a temperature of at least 33 C at the recording site. In NMJ disorders, a cold limb may result in an attenuated CMAP decrement, masking a potential CMAP decrement and resulting in a normal RNS study. Although the reason for this effect is not completely
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
197
Fig. 6. Immobilization during RNS of limb muscle. Set-up for ulnar nerve repetitive stimulation. Recording electrodes are secured with tape over the abductor digiti minimi. The stimulator is secured to the wrist with a Velcro strap or tape. The entire forearm and hand are secured to an arm board with additional Velcro straps, and the fingers are taped together. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
understood, it may be related to decreased functioning of acetylcholinesterase in the cold, effectively making more ACh available to bind at receptors. Clinically, patients with myasthenia gravis note worsening of their symptoms in warm weather, perhaps because the acetylcholinesterase is more active.
Fig. 7. Artifactual increment with submaximal stimuli. CMAP increment with 3-Hz RNS in a normal subject caused by submaximal stimulation. Note that with supramaximal stimulation, there is no increment. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
198
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Acetylcholinesterase inhibitors should be discontinued temporarily before the study If there is no medical contraindication, patients should be advised to refrain from taking acetylcholinesterase inhibitors (eg, Mestinon) for a day before RNS. Because these agents make more ACh available to bind at receptors, they may diminish a potential CMAP decrement, resulting in a normal RNS study. Nerve selection Repetitive nerve stimulation and exercise testing can be performed using any motor nerve. The most commonly used nerves for RNS include the ulnar, median, spinal accessory, musculocutaneous, axillary, and facial. In patients with suspected disorders of skeletal muscle membrane excitability, exercise testing is easily performed using the ulnar or median nerve, recording a distal muscle. In patients with postsynaptic NMJ disorders (eg, myasthenia gravis), weakness is manifested predominantly in ocular, bulbar, and proximal muscles [24]. RNS more frequently yields abnormal results in patients with NMJ defects when more proximal nerves are used (Fig. 8) [25]. Nevertheless, there are technical difficulties when stimulating proximal nerves. The spinal accessory nerve, recording the upper trapezius, probably poses the least technical difficulties when testing the proximal nerves (Fig. 9). Because the spinal accessory nerve is superficial, located just posterior to the sternocleidomastoid muscle, it can usually be stimulated supramaximally with 15 to 25 mA of
Fig. 8. 3-Hz RNS of proximal and distal nerves in patient with myasthenia gravis. Top trace shows normal decrement (4%) in the ulnar nerve. Bottom trace shows markedly abnormal decrement (42%) in the spinal accessory nerve. In myasthenia gravis, the yield of an abnormal decrement is greater with proximal nerves. Note also the U-shaped decrement. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: ButterworthHeinemann; 1998; with permission.)
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
199
Fig. 9. Spinal accessory nerve stimulation. The nerve is easily stimulated posterior to the sternocleidomastoid muscle with recording electrodes over the upper trapezius and shoulder. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
current [26]. Shoulder movement can be reduced by gentle but firm downward pressure on the shoulder or arm. The facial nerve or its branches also can be used for RNS, recording the nasalis, frontalis, orbicularis oculus, or other facial muscles; however, two problems must be kept in mind. First, facial CMAP amplitudes are small at baseline, and, second, one cannot immobilize the muscle to prevent possible electrode movement. If a facial muscle has a baseline CMAP amplitude of 1 mV, a 0.1-mV drop with RNS will result in a 10% decrement, which is generally considered significant. A 0.1-mV drop could easily occur as a result of technical factors, such as electrode movement, resulting in a spurious finding of a significant decrement where one does not exist. In contrast, the ulnar nerve may have a 10-mV CMAP baseline amplitude, which would require a 1-mV drop to yield a 10% decrement. A small change from the baseline CMAP resulting from electrode movement or the failure to perform supramaximal stimulation is much more likely to confound facial RNS, creating possible false-positive results. Stimulation frequency The optimal frequency for slow RNS is 2 or 3 Hz. The frequency must be kept slow enough to prevent calcium accumulation but fast enough to deplete the quanta in the primary store before the secondary store starts to replenish it. For rapid RNS, the optimal frequency is 30 to 50 Hz. It is always preferable to have a cooperative patient perform brief intense exercise for 10 seconds rather than performing rapid RNS, which is much more painful. Rapid RNS
200
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
should be reserved for patients who cannot cooperate with exercise, such as patients who are comatose, patients who are too weak to perform intense exercise, or infants. Number of stimulations A train of five to ten pulses is preferable for slow RNS. The number should be kept to a minimum for patient comfort but must be counterbalanced by the need to have enough pulses to detect a decrement. When the secondary store begins to resupply the primary store, the decrement begins to improve. The result is a so-called ‘‘U-shaped’’ decrement, which is highly characteristic of true NMJ disorders (see Fig. 3A). For rapid RNS, which should only be performed in patients who cannot perform brief maximal voluntary exercise, a stimulus train of 5 to 10 seconds should be given. This length of time is required to see a maximal incremental response from increased mobilization of quanta and calcium accumulation [4]. Decrement and increment calculation in repetitive nerve stimulation The decrement in RNS is usually calculated by comparing the lowest CMAP amplitude or area with the baseline CMAP amplitude (first CMAP of the RNS train). With 3-Hz RNS, the lowest CMAP amplitude is usually the third or fourth. By the fifth or sixth stimulation, the decrement begins to improve owing to the secondary store resupplying the primary store (ie, the U-shaped decrement). The CMAP decrement is expressed in a percentage and is calculated as follows: Decrement ¼
Baseline Amplitude ð1st responseÞ Lowest Amplitude ð3rd or 4th responseÞ 100 Baseline Amplitude ð1st responseÞ
Any decrement greater than 10% is considered abnormal. Theoretically, normal subjects should have no decrement; however, the 10% cutoff allows for technical factors often encountered during RNS. Any reproducible decrement is probably abnormal. Increments in rapid RNS and brief (10 seconds) intense exercise are calculated by comparing the highest CMAP amplitude or area with the baseline CMAP. The increment is expressed in a percentage and is calculated as follows: Increment ¼
Highest Amplitude Baseline Amplitude ð1st responseÞ 100 Baseline Amplitude ð1st responseÞ
With rapid RNS, the highest amplitude is usually the last one obtained after several seconds; with brief exercise, the highest amplitude is the one obtained immediately after brief exercise. In normal subjects, pseudofacilitation may cause an increment up to 40%. Increments greater than 100% are considered abnormal and are often encountered in patients with presynaptic NMJ disorders. Increments between 40% and 100% are best considered
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
201
equivocal. The increment in Lambert-Eaton myasthenic syndrome is frequently higher than 200% and more pronounced than in patients with botulism. A decremental response with RNS occurs predominantly in primary disorders of the NMJ. Nevertheless, a decrement may be seen in other disorders, especially in severe denervating disorders such as motor neuron disease [27]. In any condition in which there is prominent denervation and reinnervation, newly formed NMJs, which arise as denervated fibers are reinnervated, are initially immature and unstable. These immature and unstable NMJs may show a decrement to RNS. In addition to denervating disorders, some of the disorders of skeletal muscle membrane excitability may show a decrement to RNS [28]. Although decrements may be seen with slow RNS (2–3 Hz), they are generally more common with faster frequencies (>10 Hz) in disorders of skeletal muscle membrane excitability. Not all patients show abnormalities, and, when present, they are not specific to any individual syndrome. In these patients, additional short or prolonged exercise testing may be useful. RNS may also reveal abnormal findings in some of the metabolic myopathies (eg, McArdle’s disease), and RNS should not be performed in isolation. A clinical history and neurologic examination, routine nerve conduction studies, and routine needle EMG should be performed in every patient so that if a decremental response occurs with RNS, it can be interpreted in the proper context.
Repetitive nerve stimulation protocol The recommended RNS protocol is performed in the following manner: The extremity is warmed to 33 C. The muscle to be recorded from is immobilized as best as possible. Routine motor nerve conduction studies are performed first to ensure that the nerve is normal. Repetitive nerve stimulation is performed with the muscle at rest. After ensuring that the stimulus is supramaximal, 3-Hz RNS is performed at rest using five to ten impulses. The stimulation is repeated three times, 1 minute apart, to ensure that any decrement that is seen is reproducible. Normally, there is a less than 10% decrement between the first and fifth response. If there is a greater than 10% decrement, which is consistently reproducible, the following steps should be performed: (1) Have the patient perform maximal voluntary exercise for 10 seconds. (2) Immediately after the 10 seconds of exercise, repeat 3-Hz RNS postexercise to demonstrate postexercise facilitation and repair of the decrement. If there is less than a 10% decrement or no decrement, the following steps should be performed: (1) Have the patient perform maximal voluntary
202
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
exercise for 1 minute, followed by 3-Hz RNS, which is performed immediately and at 1, 2, 3, and 4 minutes after exercise to demonstrate postexercise exhaustion and bring out a decrement. (2) If there is greater than a 10% decrement after a minute of exercise, have the patient perform maximal voluntary exercise again for 10 seconds, followed immediately by 3-Hz RNS to demonstrate repair of the decrement. Perform RNS on one distal and one proximal motor nerve. Try to study weak muscles. If no significant decrement is found with a proximal limb muscle, a facial muscle can be tested, keeping in mind technical considerations. If the CMAP amplitude is low at baseline, have the patient perform 10 seconds of maximal voluntary exercise followed immediately by one supramaximal stimulus to look for an abnormal increment (>100% of baseline). If the patient exercises for more than 10 seconds, or if the nerve is not stimulated immediately postexercise, a potential increment may be missed. Always perform concentric needle EMG of proximal and distal muscles, especially clinically weak muscles. Any muscle with denervation or myotonia on needle EMG may demonstrate a decrement on slow RNS, which does not signify a primary disorder of the NMJ [28]. Because technical factors can complicate RNS, one must constantly determine whether the decrement makes sense in terms of NMJ physiology [8]. The following questions should be kept in mind: Is the baseline CMAP stable? If there is a CMAP decrement or increment, is it reproducible? If there is a CMAP decrement, does it repair with 10 seconds of maximal voluntary exercise (ie, postexercise facilitation)? If there is a CMAP decrement, does it worsen several minutes after extended (1 minute) exercise (postexercise exhaustion)? If the decrement worsens, can it be repaired after brief intense exercise (postexercise facilitation)? Is there a U-shaped decrement (ie, does the CMAP amplitude increase toward baseline by the fifth or sixth CMAP as mobilization of ACh increases)? If all of these criteria can be affirmed, the decrement or increment is most likely secondary to a true NMJ disorder.
Short and prolonged exercise testing in the electromyography laboratory Short and prolonged exercise testing have an important role in the EMG laboratory in the evaluation of patients with suspected disorders of skeletal muscle membrane excitability, which result from mutations of specific ion
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
203
channels or a protein kinase defect. The most common of these disorders include myotonia congenita, paramyotonia congenita, hyperkalemic periodic paralysis, hypokalemic periodic paralysis, and myotonic dystrophy. Exercise testing results in the repetitive firing of motor units, which, in some of these disorders, may lead to muscle membrane inexcitability and a progressive decline in the CMAP, associated with weakness. The pattern of decremental responses that emerges may help diagnose and differentiate among these disorders. Short exercise test In the short exercise test [13], a routine distal CMAP is evoked with supramaximal stimulation (eg, stimulating the ulnar nerve at the wrist, recording the abductor digiti minimi). The nerve is then stimulated at 1-minute intervals for several minutes to ensure a stable baseline. The patient is asked to perform maximal voluntary contraction for 5 to 10 seconds. Immediately afterward, a supramaximal CMAP is evoked. If a decrement in amplitude is seen, a CMAP is recorded every 10 seconds until the CMAP recovers to baseline (typically 1–2 minutes). If a decrement occurs after brief exercise and then recovers over 1 to 2 minutes, the same procedure is repeated several times to determine whether the decrement continues to occur or habituates. Prolonged exercise test In the prolonged exercise test, the recording procedure is the same as for the short exercise test; however, after ensuring a stable baseline, the patient is asked to contract the muscle voluntarily and maximally for 3 to 5 minutes, resting for a few seconds every 15 seconds. After the 3 to 5 minutes of exercise are completed, the patient relaxes completely. A supramaximal CMAP is evoked immediately and then every 1 to 2 minutes for the next 40 to 60 minutes. Patterns of responses in clinical disorders In the dominant and recessive forms of myotonia congenita associated with chloride channel mutations, the short exercise test produces a drop in CMAP amplitude immediately following exercise, which recovers over 1 to 2 minutes with repeated recording of the CMAP every 10 seconds (Fig. 10) [13]. In the recessive form, the initial drop in amplitude is often profound and is more likely to continue to show an initial decremental response on repeated trials of the short exercise test. In contrast, the short exercise test in myotonic dystrophy type I, associated with a protein kinase defect, produces a drop in the CMAP amplitude immediately after exercise, which recovers to baseline over 2 minutes. If short exercise is then repeated, the decremental response habituates after one or two cycles. In paramyotonia congenita, which is associated with a sodium channel mutation, the short
204
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 10. Short exercise test in the myotonic syndromes. Following a brief maximal voluntary contraction, the CMAP immediately decrements in the myotonic syndromes. If subsequent CMAPs are evoked every 10 seconds, the decrement recovers to baseline in 1 to 2 minutes in myotonic dystrophy and myotonia congenita. Numbers on the left refer to the time in seconds following the exercise. In paramyotonia congenita, the recovery may be delayed, in the range of 10 to 60 minutes. (From McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10; reprinted by permission of John Wiley & Sons, Inc.)
exercise test may produce a drop in CMAP amplitude. The drop shows a marked delay in recovery to the baseline CMAP amplitude with repeated recording of the CMAP up to an hour if the test is performed when the muscle is cooled [13]. The short exercise test produces no decrement in the periodic paralysis syndromes. In contrast, in hyperkalemic and hypokalemic periodic paralysis, the prolonged exercise test often produces an immediate increase in the CMAP amplitude, especially if the initial amplitude is low. This increase is followed by a progressive drop in the CMAP amplitude by about 50% over 20 to 40 minutes, with most of the decline occurring in the first 20 minutes following prolonged exercise (Fig. 11) [11,12]. Summary Repetitive nerve stimulation and exercise testing are useful in the evaluation of patients with suspected disorders of the NMJ and muscle membrane excitability when performed with close attention to technical factors. They can be very helpful in the diagnosis of myasthenia gravis, Lambert-Eaton
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
205
Fig. 11. Typical pattern of response on prolonged exercise test in periodic paralysis. Following 3 to 5 minutes of prolonged exercise, the CMAP amplitude recorded every 1 to 2 minutes shows little change in normal controls (A). In the periodic paralysis syndromes (B), there is frequently an increment immediately after exercise, followed by a slow decrement over the next 20 to 30 minutes. Decrements greater than 40% are definitely abnormal. (From McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10; reprinted by permission of John Wiley & Sons, Inc.)
myasthenic syndrome, and botulism, as well as rare disorders of skeletal muscle membrane excitability, including paramyotonia congenita, myotonia congenita, myotonic dystrophy, and the periodic paralyses. References [1] Keesey JC. AAEM minimonograph no. 33: electrodiagnostic approach to defects of neuromuscular transmission. Muscle Nerve 1989;12(8):613–26.
206
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
[2] Kimura J. Electrodiagnosis in diseases of nerve and muscle. Philadelphia: FA Davis; 1989. [3] Oh SJ. Botulism: electrophysiological studies. Ann Neurol 1977;1(5):481–5. [4] Oh SJ. Diverse electrophysiological spectrum of the Lambert-Eaton myasthenic syndrome. Muscle Nerve 1989;12(6):464–9. [5] Preston DC, Shapiro BE. Repetitive nerve stimulation. In: Preston DC, Shapiro BE, editors. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998. [6] Ptacek LJ, Johnson KJ, Griggs RC. Genetics and physiology of the myotonic muscle disorders. N Engl J Med 1993;18:482–9. [7] Shapiro BE, Ruff RL. Disorders of skeletal muscle membrane excitability: myotonia congenita, paramyotonia congenita, periodic paralysis and related syndromes. In: Katirji B, Kaminski H, Preston DC, Ruff RL, Shapiro BE, editors. Neuromuscular disorders in clinical practice. Woburn (MA): Butterworth-Heinemann; 2002. [8] AAEM Professional Practice Committee, Chiou-Tan FY, Gilchrist JM, Tim RW. Literature review of the usefulness of repetitive nerve stimulation and single fiber EMG in the electrodiagnostic evaluation of patients with suspected myasthenia gravis or Lambert-Eaton myasthenic syndrome. Muscle Nerve 2001;24:1239–47. [9] Jablecki C. AAEM case report no. 3: myasthenia gravis. Muscle Nerve 1991;14(5):391–7. [10] Tim RW, Sanders DB. Repetitive nerve stimulation studies in the Lambert-Eaton myasthenic syndrome. Muscle Nerve 1994;17(9):995–1001. [11] Kuntzer T, Flocard F, Vial C, Kohler A. Exercise test in muscle channelopathies and other muscle disorders. Muscle Nerve 2000;23:1089–94. [12] McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10. [13] Streib EW. AAEE minimononograph no. 27: differential diagnosis of myotonic syndromes. Muscle Nerve 1987;10:603–15. [14] Hubbard JI. Microphysiology of vertebrate neuromuscular transmission. Physiol Rev 1973;53:674–723. [15] Slater CR, Lyons PR, Walls TH, Fawcett PRW, Young C. Structure and function of neuromuscular junctions in the vastus lateralis of man. Brain 1992;11:337–48. [16] Fatt P, Katz B. Spontaneous subthreshold activity at motor nerve endings. J Physiol 1952;117:109–28. [17] Paton WD, Waud DR. The margin of safety of neuromuscular transmission. J Physiol 1967;191:59–90. [18] Engel AG. Myasthenia gravis and myasthenic disorders. New York: Oxford University Press; 1999. [19] Lindstrom JM. Acetylcholine receptors and myasthenia. Muscle Nerve 2000;23:453–77. [20] Engel AG. Lambert-Eaton myasthenic syndrome. Ann Neurol 1987;22:193–9. [21] Lambert EH, Elmqvist D. Quantal components of end-plate potentials in the myasthenic syndrome. Ann N Y Acad Sci 1971;183:183–99. [22] Jablecki CK. Electrodiagnostic evaluation of patients with myasthenia gravis and related disorders. Neurol Clin 1985;3(3):557–72. [23] Tim RW, Massey JM, Sanders DB. Lambert-Eaton myasthenic syndrome: electrodiagnostic findings and response to treatment. Neurology 2000;54:2176–8. [24] Jablecki C. Lambert-Eaton myasthenic syndrome. Muscle Nerve 1984;7:250–7. [25] Schady W, MacDermott N. On the choice of muscle in the electrophysiological assessment of myasthenia gravis. Electromyogr Clin Neurophysiol 1992;32(3):99–102. [26] Schumm F, Stohr M. Accessory nerve stimulation in the assessment of myasthenia gravis. Muscle Nerve 1984;7(2):147–51. [27] Bernstein LP, Antel JP. Motor neuron disease: decremental responses to repetitive nerve stimulation. Neurology 1981;31(2):204–7. [28] Aminoff MJ, Layzer RB, Satya-Murti S, Faden AI. The declining electrical response of muscle to repetitive nerve stimulation in myotonia. Neurology 1977;27:812–6.
Repetitive nerve stimulation and exercise testing Barbara E. Shapiro, MD, PhD*, David C. Preston, MD Department of Neurology, University Hospitals of Cleveland, Case Western Reserve University, 11100 Euclid Avenue, Cleveland, OH 44106-5098, USA
Repetitive nerve stimulation (RNS) and exercise testing have a key role in the electrodiagnostic evaluation of patients with suspected neuromuscular junction (NMJ) disorders, especially myasthenia gravis, Lambert-Eaton myasthenic syndrome, and botulism [1–5]. In addition, these studies are often useful in patients with rare disorders of skeletal muscle membrane excitability, including paramyotonia congenita, myotonia congenita, myotonic dystrophy, and the periodic paralyses [6,7]. The technique dates back to the late 1800s when Jolly first made visual observations of muscle movement following RNS. Unlike contemporary methods, his initial studies were performed with submaximal stimuli, using mechanical rather than electrical measurements. Using these techniques, Jolly noted a decremental response following RNS in patients with myasthenia gravis and correctly concluded that the disorder had a peripheral rather than central origin. Since that time, RNS and exercise testing have been advanced and validated to become one of the most useful tests in the evaluation of patients with disorders of the NMJ and muscle membrane excitability [8,9]. In any patient presenting with symptoms of a possible NMJ disorder (eg, fatigability, proximal weakness, dysphagia, dysarthria, or ocular abnormalities), one should perform RNS as part of the electrodiagnostic evaluation. The pattern of responses to slow (2–3 Hz) and rapid RNS (30–50 Hz) is useful in determining whether a NMJ disorder is present and whether it is presynaptic or postsynaptic in origin [5,8]. In addition to RNS, exercise testing has an important role in the evaluation of patients with suspected NMJ disorders. Abnormalities on slow RNS are often more pronounced several minutes following exercise testing and can increase the sensitivity of the study. * Corresponding author. E-mail address: [email protected] (B.E. Shapiro). 1047-9651/03/$ – see front matter Ó 2003, Elsevier Inc. All rights reserved. doi:10.1016/S1047-9651(02)00129-8
186
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Brief maximal voluntary exercise can be used to demonstrate many of the same effects as rapid RNS, which are often key in detecting a presynaptic NMJ disorder. Rapid RNS is painful and difficult to tolerate, whereas exercise testing has the distinct advantage of being painless and is preferable to rapid RNS in patients who are alert and cooperative [10]. When a cooperative subject is asked to contract a muscle voluntarily at maximum force, motor units fire at their maximal firing frequency, typically 30 to 50 Hz. Brief maximal voluntary exercise can be used as a painless substitute to demonstrate the same effects as rapid RNS. Less commonly, exercise testing is employed in the electrodiagnostic evaluation of the rare patient with a suspected disorder of muscle membrane excitability. Exercise testing can be performed over a short (ie, 10 seconds) or prolonged (ie, several minutes) period. In some of these disorders, muscle membrane inexcitability may follow exercise testing, with a corresponding and progressive decline in the compound muscle action potential (CMAP) amplitude as weakness develops [7,11,12]. Exercise testing is useful not only in detecting a muscle membrane excitability disorder but can often aid in further narrowing the differential diagnosis based on the pattern of abnormalities [13]. In the electromyography (EMG) laboratory, the effects of RNS and exercise testing on the CMAP are studied. Analysis of any decremental or incremental response forms the basis of the study. To interpret these responses accurately, one must understand normal NMJ physiology and the effects of RNS and exercise on a single NMJ and its associated muscle fiber. This knowledge can then be used in the EMG laboratory to predict accurately the effect of RNS and exercise testing on the CMAP in normal subjects and in patients with NMJ and skeletal muscle membrane excitability disorders.
Normal neuromuscular junction physiology The NMJ forms an electrical-chemical-electrical link between nerve and muscle [14] (Fig. 1). The chemical neurotransmitter at the NMJ is acetylcholine (ACh). ACh molecules are packaged as vesicles in discrete units known as quanta within the presynaptic terminal. Each quantum contains approximately 10,000 molecules of ACh, located in three separate stores. The primary or immediately available store consists of approximately 1000 quanta located just beneath the presynaptic nerve terminal membrane. As its name implies, this store is immediately available for release. The secondary or mobilization store consists of approximately 10,000 quanta that can resupply the primary store after a few seconds. A tertiary or reserve store of more than 100,000 quanta is located far from the NMJ in the axon and cell body. When a nerve action potential invades and depolarizes the presynaptic junction, voltage-dependent calcium channels are activated, allowing an
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
187
Fig. 1. Normal NMJ anatomy. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
influx of calcium. This infusion of calcium results in the release of ACh from the presynaptic terminal. The greater the calcium concentration inside the presynaptic terminal, the more quanta are released. ACh then diffuses across the synaptic cleft and binds to ACh receptors on the postsynaptic muscle membrane. The postsynaptic membrane is composed of numerous junctional folds, effectively increasing the surface area of the membrane, with ACh receptors clustered on the crests of the folds [15]. The binding of ACh to its receptors opens ion channels, resulting in a local depolarization, the end-plate potential (EPP) [16]. The size of the EPP is proportional to the amount of ACh that binds to the receptors. If the EPP depolarizes the muscle membrane above threshold, an allor-none muscle fiber action potential is generated and propagated through the muscle fiber, similar to the generation of an all-or-none nerve action potential. Under normal circumstances, the EPP always rises above threshold, resulting in a muscle fiber action potential. The amplitude of the EPP above the threshold value needed to generate a muscle fiber action potential is called the safety factor [17]. ACh is then broken down in the synaptic cleft by the enzyme
188
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
acetylcholinesterase, with the choline subsequently taken back up into the presynaptic terminal to be repackaged into ACh. During slow RNS (2–3 Hz), ACh quanta are progressively depleted from the primary store, and subsequently fewer quanta are released with each successive stimulation. The corresponding EPP decreases in amplitude. In normal subjects, because of the normal safety factor, the EPP remains above threshold, ensuring the generation of a muscle fiber action potential with each stimulation. After the first few seconds, the secondary store begins to replace the depleted quanta with a subsequent rise in the EPP. The physiology of rapid RNS (30–50 Hz) in normal subjects is more complicated. Depletion of quanta from the presynaptic terminal is counterbalanced not only by the mobilization of quanta from the secondary store but also by the accumulation of calcium. Normally, approximately 100 ms are required for calcium to diffuse back out of the presynaptic terminal. If RNS is rapid enough so that new calcium infuses in before the previously infused calcium has diffused back out, calcium continues to accumulate in the presynaptic terminal, resulting in an increased release of quanta. This increased number of quanta results in a correspondingly higher EPP. Nevertheless, in normal subjects, the result is the same as with any other EPP above threshold, that is, an all-or-none muscle fiber action potential. Although the effects of slow and rapid RNS are very different at the molecular level, in normal subjects, the result is the same—the consistent generation of a muscle fiber action potential. In pathologic conditions in which the safety factor is reduced (ie, the baseline EPP is reduced but still above threshold), slow RNS will cause depletion of quanta and may drop the EPP below threshold, resulting in the absence of a muscle fiber action potential. In pathologic conditions in which the baseline EPP is below threshold and a muscle fiber action potential is not generated, rapid RNS may increase the number of quanta released, resulting in a larger EPP such that threshold is reached. Once threshold is reached, a muscle fiber action potential is generated where one had not been present previously. These concepts form the basis of decrements to slow RNS and increments to rapid RNS and brief exercise seen in patients with NMJ disorders.
Repetitive nerve stimulation modeled at the level of the neuromuscular junction and its associated muscle fiber Repetitive nerve stimulation in normal subjects and patients with NMJ disorders can be modeled effectively by making the following three assumptions: m ¼ pn, where m is the number of quanta released during each stimulation; p is the probability of release (effectively proportional to the concentration of calcium), typically approximately 0.2 in normals; and n is the number of quanta in the primary store (at baseline, typically approximately 1000 in normals).
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
189
The secondary store starts to replenish the primary store after 1 to 2 seconds. Calcium requires approximately 100 ms to be pumped out of the presynaptic terminal. If stimulation occurs again before 100 ms (ie, stimulation rate >10 Hz), the calcium concentration increases, the probability of release of ACh quanta increases, and more quanta are released. Slow repetitive nerve stimulation In normal subjects during slow RNS, approximately 20% of the quanta are released from the primary store in the first stimulation and with each subsequent stimulation. If the resultant EPP is above threshold, a muscle fiber action potential is generated. Between the first and fourth stimulation, there is a normal depletion of the primary store, a subsequent decline in the number of quanta released, and a corresponding decrease in the EPP. By the fifth stimulation, sufficient time has elapsed for the secondary store to begin to resupply the primary store. As the number of quanta in the primary store increases at this point, with a corresponding increase in the number of ACh quanta released, a higher EPP is generated. In normal subjects, because the EPP stays above threshold at all times, the generation of a muscle fiber action potential is consistent (Fig. 2A). In contrast, in postsynaptic NMJ disorders, the same number of quanta generates a smaller EPP. Accordingly, the safety factor is reduced. In myasthenia gravis, the most common of the postsynaptic disorders, there are fewer ACh receptors and, accordingly, less binding of ACh [18,19]. The reduced safety factor in conjunction with the normal depletion of quanta results in subsequent EPPs falling below threshold and their corresponding muscle fiber action potentials not being generated (Fig. 2B). In Lambert-Eaton myasthenic syndrome, a presynaptic disorder, the number of quanta in the primary store is normal, and the EPP is normal for the number of quanta released. The abnormalities include the reduced number of ACh quanta released and the low baseline EPP [20,21]. The calcium concentration in the presynaptic terminal is reduced owing to an antibody attack on the voltage-gated calcium channels. The probability of release dramatically falls, with a decrease in the number of quanta released. There is still depletion, although not as marked as in normal subjects or patients with postsynaptic disorders, simply because so few quanta are released that the subsequent amount of depletion cannot be as great. Because the EPP is below threshold at baseline, a muscle fiber action potential is never generated (Fig. 2C). In some presynaptic disorders, the baseline EPP may be low but still above threshold, resulting in a reduced safety factor. In this situation, a muscle fiber action potential may be generated initially but then fails to be generated as the EPP falls below threshold with slow RNS.
190
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 2. End-plate potentials. Dotted line (- - - -) marks the threshold. Shaded EPPs rise above threshold and generate a muscle fiber action potential. (A) 3-Hz RNS, normal NMJ. Note that all of the potentials remain well above threshold despite the normal decline in EPP amplitude (safety factor). (B) 3-Hz RNS, postsynaptic NMJ disorder. Note the lower EPP amplitudes. With further ACh depletion, the last three potentials fall below threshold, and a muscle fiber action potential is not generated. (C ) 3-Hz RNS, presynaptic NMJ disorder. Note that all of the EPPs are below threshold, and no muscle fiber action potentials are generated. The EPP declines in amplitude but not as markedly as in normal subjects or patients with postsynaptic NMJ disorders. (D) 50-Hz RNS, presynaptic NMJ disorder. Note the progressive increment in the EPP amplitude to above threshold and the subsequent generation of muscle fiber action potentials. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
Rapid repetitive nerve stimulation The effects of rapid RNS can be deduced from the three basic assumptions described previously. With rapid RNS, the depletion of quanta is counterbalanced by (1) increased mobilization of quanta from the secondary to the primary store and (2) calcium accumulation in the presynaptic terminal,
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
191
which increases the probability of release. The sum of these influences usually results in a greater number of quanta released and higher EPPs with RNS. In normal subjects, rapid RNS always results in the generation of a muscle fiber action potential, the same as with any EPP above threshold. In patients with postsynaptic NMJ disorders, the EPP will also increase, but, because the EPP is usually above threshold at baseline, the result will still be the generation of a muscle fiber action potential. If the EPP has been lowered, such as after slow RNS, the decreased EPP may be repaired or improved with rapid RNS. If the EPP has dropped below threshold, subsequent rapid RNS may increase the EPP back to above threshold. In patients with presynaptic NMJ disorders, the situation is distinctly different. Because the EPP is abnormally low at baseline and often below threshold, rapid RNS may increase the EPP above threshold so that a muscle fiber action potential is generated where one had not been present previously (Fig. 2D). Repetitive nerve stimulation and brief exercise testing in the electromyography laboratory The previous discussion has pertained to end-plate and individual muscle fiber action potentials. During RNS in the EMG laboratory, all measurements are made on the CMAP, the sum of the individual muscle fiber action potentials generated in a muscle. One infers that the CMAP amplitude and area are proportional to the number of muscle fibers activated. In normal subjects, although the EPP is affected by slow and rapid RNS, the potential always remains above threshold, resulting in the consistent generation of muscle fiber action potentials. In normal subjects, CMAPs generated following either slow or rapid RNS do not change significantly in amplitude or area. Slow repetitive nerve stimulation In patients with NMJ disorders, if the normal EPP safety factor is reduced, slow RNS will result in the depletion of quanta and reduce the amplitude of the EPP. If the EPP of some muscle fibers falls below threshold, those muscle fiber action potentials will not be generated, and the number of individual muscle fiber action potentials will decline. As the number of individual muscle fiber action potentials declines, a decrement of the CMAP amplitude and area likewise occurs. This effect provides the basis for a decremental CMAP response to slow RNS in the EMG laboratory, which reflects fewer EPPs reaching threshold and fewer individual muscle fiber action potentials contributing to the CMAP. Rapid repetitive nerve stimulation In patients with NMJ disorders in whom some EPPs are below threshold at baseline (usually the presynaptic NMJ disorders), rapid RNS can be used to
192
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
facilitate the EPP. If subthreshold EPPs can be brought above threshold, muscle fiber action potentials will be generated where they had not been present previously, and the number of individual muscle fiber action potentials will increase. This effect provides the basis for an incremental CMAP response to rapid RNS seen in the EMG laboratory. As the number of individual muscle fiber action potentials increases, an increment of the CMAP amplitude and area occurs, which reflects more EPPs reaching threshold and more individual muscle fiber action potentials contributing to the CMAP. Incremental responses greater than 100% above baseline in response to rapid RNS are not unusual in presynaptic NMJ disorders. Brief exercise testing In postsynaptic NMJ disorders, brief exercise, just like rapid RNS, results in higher EPPs; however, because the EPP is usually above threshold at baseline, the result is the same, that is, the generation of a muscle fiber action potential. Brief exercise may repair or improve a low EPP that has developed following slow RNS. If the EPP has dropped below threshold, subsequent exercise may increase the EPP back to above threshold. In presynaptic NMJ disorders, brief exercise, like rapid RNS, can often facilitate low EPPs. If the baseline EPP is below threshold, brief exercise may increase the EPP above threshold so that a muscle fiber action potential is generated where one had not been present previously. As is true with rapid RNS, as the number of individual muscle fiber action potentials increases, brief exercise results in an increment of the CMAP amplitude and area. This increment reflects more EPPs reaching threshold and more individual muscle fiber action potentials contributing to the CMAP. Post exercise facilitation and exhaustion The effects of rapid RNS or voluntary exercise occur with brief periods of exercise or rapid RNS, typically 10 seconds. This process is known as postexercise (or posttetanic) facilitation. A phenomenon of postexercise (or posttetanic) exhaustion, which is not as well understood, occurs after a more extended period of exercise. Immediately after an extended period of exercise or rapid RNS (usually 1 minute), EPPs typically increase initially, followed by a subsequent decline in EPPs, continuing and worsening over the next several minutes. In normal subjects, the EPP never falls below threshold because of the normal safety factor. In contrast, in patients with impaired NMJ transmission, when slow RNS is performed 2 to 4 minutes after extended exercise, decrements are more pronounced when compared with the baseline, because many more EPPs no longer reach threshold, and their corresponding muscle fiber action potentials are not generated. The effects of postexercise facilitation and postexercise exhaustion can be demonstrated on the CMAP in patients with NMJ disorders (Fig. 3). Post-
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
193
exercise facilitation can be demonstrated in two situations. Brief exercise can repair EPPs that have been lowered by slow RNS. If the EPPs are facilitated above threshold, muscle fiber action potentials will be generated that were not present previously. Accordingly, a decrement of the CMAP amplitude and area that developed during slow RNS may be lessened or ‘‘repaired’’ (Fig. 3A, B). In presynaptic NMJ disorders such as Lambert-Eaton myasthenic syndrome associated with reduced release of quanta and subthreshold EPPs at baseline, brief exercise can facilitate EPPs above threshold, giving rise to muscle fiber action potentials that were not present previously [22,23]. Accordingly, an increment of the CMAP amplitude and area occurs (Fig. 4). To demonstrate the phenomenon of postexercise exhaustion, the muscle is maximally exercised for 1 minute. Slow RNS is then performed immediately after exercise and at 1, 2, 3, and 4 minutes later. In normal subjects with a normal safety factor, the EPP never falls below threshold, and the CMAP amplitude and area remain stable. In contrast, in patients with impaired NMJ transmission, the decrement in CMAP amplitude and area in response to slow RNS becomes more marked 2 to 4 minutes after a minute of exercise (Fig. 3C, D, E). In this situation, 10 seconds of maximal voluntary exercise will repair the decrement toward normal (postexercise facilitation) (Fig. 3F). An important additional concept known as pseudofacilitation should be appreciated by every electromyographer in regards to brief exercise testing or rapid RNS. In normal subjects, brief maximal exercise often leads to a slight increase in CMAP amplitude, which is essential not to misinterpret as an abnormality. Following brief exercise, EPPs are facilitated; however, because they are above threshold at baseline, the same number of muscle fiber action potentials is generated. Although there is no increase in the actual number of muscle fiber action potentials that summate to create the CMAP, brief maximal exercise causes the muscle fibers to fire more synchronously. Because the CMAP is composed of many summated individual muscle fibers, increased synchrony of firing results in slightly better alignment of the negative and positive phases of the individual muscle fiber action potentials, less normal phase cancellation, and a slightly higher CMAP. This pseudofacilitation results in a small increase in CMAP amplitude but usually with a small decrease in CMAP duration and little change in the CMAP area (Fig. 5). In general, postexercise increments of CMAP amplitude from pseudofacilitation do not exceed 40% of baseline in normal subjects.
Technical factors in repetitive nerve stimulation and exercise testing Close attention to technical factors is critical when performing RNS and exercise testing. Technical factors may result in spurious decrements or increments and the mistaken impression of a NMJ disorder if not carefully controlled.
194
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
195
Fig. 4. Exercise testing in Lambert-Eaton myasthenic syndrome. Stimulating the median nerve supramaximally at the wrist, recording from the abductor pollicis brevis muscle. Top trace shows baseline. Bottom trace shows response immediately after 10 seconds of maximal voluntary exercise. Note the marked increase in CMAP amplitude (postexercise facilitation). Preexercise and postexercise testing, looking for an increment, is always better tolerated by patients than is 50-Hz RNS. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
Immobilization: isometric electrode position is essential The greatest technical difficulty with RNS and exercise testing is the need to immobilize the recording electrode properly over the muscle. If the position of the recording electrode moves in relationship to the muscle during stimulation, the CMAP configuration may change. The goal is to minimize any movement of the limb, stimulator, or recording electrodes during RNS and
b Fig. 3. Postexercise facilitation and exhaustion. 3-Hz RNS in a patient with myasthenia gravis. (A) Decrement of CMAP amplitude at rest. (B) Postexercise facilitation. Decrement of CMAP occurs immediately following 10 seconds of maximal voluntary exercise and subsequently repairs toward normal. (C, D, E) Postexercise exhaustion. Decrements of CMAP at 1, 2, and 3 minutes after 1 minute of maximal voluntary exercise. The decrement becomes progressively more marked over the baseline decrement. (F ) Postexercise facilitation after a decrement. Immediately following another 10 seconds of maximal voluntary exercise, the decrement, which has worsened as a result of postexercise exhaustion, repairs toward normal. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
196
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 5. Pseudofacilitation. When performing RNS following exercise testing, pseudofacilitation is often encountered. Pseudofacilitation is a normal phenomena caused by more synchronous firing of muscle fiber action potentials immediately following brief intense exercise. In the normal subject, 3-Hz RNS results in a 0% CMAP decrement at rest (top trace). Immediately following 10 seconds of maximal voluntary exercise, 3-Hz RNS is repeated (bottom trace). A similar 0% decrement is found; however, the CMAP amplitudes are higher, the durations are shorter, and the areas are similar from the normal effects of pseudofacilitation. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: ButterworthHeinemann; 1998; with permission.)
exercise testing. The recording electrodes should be well secured with tape. If possible, the stimulator should be secured with tape or a Velcro strap and the entire limb secured to a pad or board (Fig. 6). Immobilization is more easily accomplished when stimulating distal nerves such as the median or ulnar. Securing the stimulator and limb is more problematic when testing proximal nerves. Stimuli must be supramaximal The use of submaximal stimulation can create a variety of problems, including spurious CMAP decrements and increments (Fig. 7). A supramaximal stimulus must always be obtained before beginning RNS and exercise testing. Temperature control is essential Repetitive nerve stimulation and exercise testing in the EMG laboratory should be performed with the limb at a temperature of at least 33 C at the recording site. In NMJ disorders, a cold limb may result in an attenuated CMAP decrement, masking a potential CMAP decrement and resulting in a normal RNS study. Although the reason for this effect is not completely
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
197
Fig. 6. Immobilization during RNS of limb muscle. Set-up for ulnar nerve repetitive stimulation. Recording electrodes are secured with tape over the abductor digiti minimi. The stimulator is secured to the wrist with a Velcro strap or tape. The entire forearm and hand are secured to an arm board with additional Velcro straps, and the fingers are taped together. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
understood, it may be related to decreased functioning of acetylcholinesterase in the cold, effectively making more ACh available to bind at receptors. Clinically, patients with myasthenia gravis note worsening of their symptoms in warm weather, perhaps because the acetylcholinesterase is more active.
Fig. 7. Artifactual increment with submaximal stimuli. CMAP increment with 3-Hz RNS in a normal subject caused by submaximal stimulation. Note that with supramaximal stimulation, there is no increment. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
198
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Acetylcholinesterase inhibitors should be discontinued temporarily before the study If there is no medical contraindication, patients should be advised to refrain from taking acetylcholinesterase inhibitors (eg, Mestinon) for a day before RNS. Because these agents make more ACh available to bind at receptors, they may diminish a potential CMAP decrement, resulting in a normal RNS study. Nerve selection Repetitive nerve stimulation and exercise testing can be performed using any motor nerve. The most commonly used nerves for RNS include the ulnar, median, spinal accessory, musculocutaneous, axillary, and facial. In patients with suspected disorders of skeletal muscle membrane excitability, exercise testing is easily performed using the ulnar or median nerve, recording a distal muscle. In patients with postsynaptic NMJ disorders (eg, myasthenia gravis), weakness is manifested predominantly in ocular, bulbar, and proximal muscles [24]. RNS more frequently yields abnormal results in patients with NMJ defects when more proximal nerves are used (Fig. 8) [25]. Nevertheless, there are technical difficulties when stimulating proximal nerves. The spinal accessory nerve, recording the upper trapezius, probably poses the least technical difficulties when testing the proximal nerves (Fig. 9). Because the spinal accessory nerve is superficial, located just posterior to the sternocleidomastoid muscle, it can usually be stimulated supramaximally with 15 to 25 mA of
Fig. 8. 3-Hz RNS of proximal and distal nerves in patient with myasthenia gravis. Top trace shows normal decrement (4%) in the ulnar nerve. Bottom trace shows markedly abnormal decrement (42%) in the spinal accessory nerve. In myasthenia gravis, the yield of an abnormal decrement is greater with proximal nerves. Note also the U-shaped decrement. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: ButterworthHeinemann; 1998; with permission.)
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
199
Fig. 9. Spinal accessory nerve stimulation. The nerve is easily stimulated posterior to the sternocleidomastoid muscle with recording electrodes over the upper trapezius and shoulder. (From Preston DC, Shapiro BE. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998; with permission.)
current [26]. Shoulder movement can be reduced by gentle but firm downward pressure on the shoulder or arm. The facial nerve or its branches also can be used for RNS, recording the nasalis, frontalis, orbicularis oculus, or other facial muscles; however, two problems must be kept in mind. First, facial CMAP amplitudes are small at baseline, and, second, one cannot immobilize the muscle to prevent possible electrode movement. If a facial muscle has a baseline CMAP amplitude of 1 mV, a 0.1-mV drop with RNS will result in a 10% decrement, which is generally considered significant. A 0.1-mV drop could easily occur as a result of technical factors, such as electrode movement, resulting in a spurious finding of a significant decrement where one does not exist. In contrast, the ulnar nerve may have a 10-mV CMAP baseline amplitude, which would require a 1-mV drop to yield a 10% decrement. A small change from the baseline CMAP resulting from electrode movement or the failure to perform supramaximal stimulation is much more likely to confound facial RNS, creating possible false-positive results. Stimulation frequency The optimal frequency for slow RNS is 2 or 3 Hz. The frequency must be kept slow enough to prevent calcium accumulation but fast enough to deplete the quanta in the primary store before the secondary store starts to replenish it. For rapid RNS, the optimal frequency is 30 to 50 Hz. It is always preferable to have a cooperative patient perform brief intense exercise for 10 seconds rather than performing rapid RNS, which is much more painful. Rapid RNS
200
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
should be reserved for patients who cannot cooperate with exercise, such as patients who are comatose, patients who are too weak to perform intense exercise, or infants. Number of stimulations A train of five to ten pulses is preferable for slow RNS. The number should be kept to a minimum for patient comfort but must be counterbalanced by the need to have enough pulses to detect a decrement. When the secondary store begins to resupply the primary store, the decrement begins to improve. The result is a so-called ‘‘U-shaped’’ decrement, which is highly characteristic of true NMJ disorders (see Fig. 3A). For rapid RNS, which should only be performed in patients who cannot perform brief maximal voluntary exercise, a stimulus train of 5 to 10 seconds should be given. This length of time is required to see a maximal incremental response from increased mobilization of quanta and calcium accumulation [4]. Decrement and increment calculation in repetitive nerve stimulation The decrement in RNS is usually calculated by comparing the lowest CMAP amplitude or area with the baseline CMAP amplitude (first CMAP of the RNS train). With 3-Hz RNS, the lowest CMAP amplitude is usually the third or fourth. By the fifth or sixth stimulation, the decrement begins to improve owing to the secondary store resupplying the primary store (ie, the U-shaped decrement). The CMAP decrement is expressed in a percentage and is calculated as follows: Decrement ¼
Baseline Amplitude ð1st responseÞ Lowest Amplitude ð3rd or 4th responseÞ 100 Baseline Amplitude ð1st responseÞ
Any decrement greater than 10% is considered abnormal. Theoretically, normal subjects should have no decrement; however, the 10% cutoff allows for technical factors often encountered during RNS. Any reproducible decrement is probably abnormal. Increments in rapid RNS and brief (10 seconds) intense exercise are calculated by comparing the highest CMAP amplitude or area with the baseline CMAP. The increment is expressed in a percentage and is calculated as follows: Increment ¼
Highest Amplitude Baseline Amplitude ð1st responseÞ 100 Baseline Amplitude ð1st responseÞ
With rapid RNS, the highest amplitude is usually the last one obtained after several seconds; with brief exercise, the highest amplitude is the one obtained immediately after brief exercise. In normal subjects, pseudofacilitation may cause an increment up to 40%. Increments greater than 100% are considered abnormal and are often encountered in patients with presynaptic NMJ disorders. Increments between 40% and 100% are best considered
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
201
equivocal. The increment in Lambert-Eaton myasthenic syndrome is frequently higher than 200% and more pronounced than in patients with botulism. A decremental response with RNS occurs predominantly in primary disorders of the NMJ. Nevertheless, a decrement may be seen in other disorders, especially in severe denervating disorders such as motor neuron disease [27]. In any condition in which there is prominent denervation and reinnervation, newly formed NMJs, which arise as denervated fibers are reinnervated, are initially immature and unstable. These immature and unstable NMJs may show a decrement to RNS. In addition to denervating disorders, some of the disorders of skeletal muscle membrane excitability may show a decrement to RNS [28]. Although decrements may be seen with slow RNS (2–3 Hz), they are generally more common with faster frequencies (>10 Hz) in disorders of skeletal muscle membrane excitability. Not all patients show abnormalities, and, when present, they are not specific to any individual syndrome. In these patients, additional short or prolonged exercise testing may be useful. RNS may also reveal abnormal findings in some of the metabolic myopathies (eg, McArdle’s disease), and RNS should not be performed in isolation. A clinical history and neurologic examination, routine nerve conduction studies, and routine needle EMG should be performed in every patient so that if a decremental response occurs with RNS, it can be interpreted in the proper context.
Repetitive nerve stimulation protocol The recommended RNS protocol is performed in the following manner: The extremity is warmed to 33 C. The muscle to be recorded from is immobilized as best as possible. Routine motor nerve conduction studies are performed first to ensure that the nerve is normal. Repetitive nerve stimulation is performed with the muscle at rest. After ensuring that the stimulus is supramaximal, 3-Hz RNS is performed at rest using five to ten impulses. The stimulation is repeated three times, 1 minute apart, to ensure that any decrement that is seen is reproducible. Normally, there is a less than 10% decrement between the first and fifth response. If there is a greater than 10% decrement, which is consistently reproducible, the following steps should be performed: (1) Have the patient perform maximal voluntary exercise for 10 seconds. (2) Immediately after the 10 seconds of exercise, repeat 3-Hz RNS postexercise to demonstrate postexercise facilitation and repair of the decrement. If there is less than a 10% decrement or no decrement, the following steps should be performed: (1) Have the patient perform maximal voluntary
202
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
exercise for 1 minute, followed by 3-Hz RNS, which is performed immediately and at 1, 2, 3, and 4 minutes after exercise to demonstrate postexercise exhaustion and bring out a decrement. (2) If there is greater than a 10% decrement after a minute of exercise, have the patient perform maximal voluntary exercise again for 10 seconds, followed immediately by 3-Hz RNS to demonstrate repair of the decrement. Perform RNS on one distal and one proximal motor nerve. Try to study weak muscles. If no significant decrement is found with a proximal limb muscle, a facial muscle can be tested, keeping in mind technical considerations. If the CMAP amplitude is low at baseline, have the patient perform 10 seconds of maximal voluntary exercise followed immediately by one supramaximal stimulus to look for an abnormal increment (>100% of baseline). If the patient exercises for more than 10 seconds, or if the nerve is not stimulated immediately postexercise, a potential increment may be missed. Always perform concentric needle EMG of proximal and distal muscles, especially clinically weak muscles. Any muscle with denervation or myotonia on needle EMG may demonstrate a decrement on slow RNS, which does not signify a primary disorder of the NMJ [28]. Because technical factors can complicate RNS, one must constantly determine whether the decrement makes sense in terms of NMJ physiology [8]. The following questions should be kept in mind: Is the baseline CMAP stable? If there is a CMAP decrement or increment, is it reproducible? If there is a CMAP decrement, does it repair with 10 seconds of maximal voluntary exercise (ie, postexercise facilitation)? If there is a CMAP decrement, does it worsen several minutes after extended (1 minute) exercise (postexercise exhaustion)? If the decrement worsens, can it be repaired after brief intense exercise (postexercise facilitation)? Is there a U-shaped decrement (ie, does the CMAP amplitude increase toward baseline by the fifth or sixth CMAP as mobilization of ACh increases)? If all of these criteria can be affirmed, the decrement or increment is most likely secondary to a true NMJ disorder.
Short and prolonged exercise testing in the electromyography laboratory Short and prolonged exercise testing have an important role in the EMG laboratory in the evaluation of patients with suspected disorders of skeletal muscle membrane excitability, which result from mutations of specific ion
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
203
channels or a protein kinase defect. The most common of these disorders include myotonia congenita, paramyotonia congenita, hyperkalemic periodic paralysis, hypokalemic periodic paralysis, and myotonic dystrophy. Exercise testing results in the repetitive firing of motor units, which, in some of these disorders, may lead to muscle membrane inexcitability and a progressive decline in the CMAP, associated with weakness. The pattern of decremental responses that emerges may help diagnose and differentiate among these disorders. Short exercise test In the short exercise test [13], a routine distal CMAP is evoked with supramaximal stimulation (eg, stimulating the ulnar nerve at the wrist, recording the abductor digiti minimi). The nerve is then stimulated at 1-minute intervals for several minutes to ensure a stable baseline. The patient is asked to perform maximal voluntary contraction for 5 to 10 seconds. Immediately afterward, a supramaximal CMAP is evoked. If a decrement in amplitude is seen, a CMAP is recorded every 10 seconds until the CMAP recovers to baseline (typically 1–2 minutes). If a decrement occurs after brief exercise and then recovers over 1 to 2 minutes, the same procedure is repeated several times to determine whether the decrement continues to occur or habituates. Prolonged exercise test In the prolonged exercise test, the recording procedure is the same as for the short exercise test; however, after ensuring a stable baseline, the patient is asked to contract the muscle voluntarily and maximally for 3 to 5 minutes, resting for a few seconds every 15 seconds. After the 3 to 5 minutes of exercise are completed, the patient relaxes completely. A supramaximal CMAP is evoked immediately and then every 1 to 2 minutes for the next 40 to 60 minutes. Patterns of responses in clinical disorders In the dominant and recessive forms of myotonia congenita associated with chloride channel mutations, the short exercise test produces a drop in CMAP amplitude immediately following exercise, which recovers over 1 to 2 minutes with repeated recording of the CMAP every 10 seconds (Fig. 10) [13]. In the recessive form, the initial drop in amplitude is often profound and is more likely to continue to show an initial decremental response on repeated trials of the short exercise test. In contrast, the short exercise test in myotonic dystrophy type I, associated with a protein kinase defect, produces a drop in the CMAP amplitude immediately after exercise, which recovers to baseline over 2 minutes. If short exercise is then repeated, the decremental response habituates after one or two cycles. In paramyotonia congenita, which is associated with a sodium channel mutation, the short
204
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
Fig. 10. Short exercise test in the myotonic syndromes. Following a brief maximal voluntary contraction, the CMAP immediately decrements in the myotonic syndromes. If subsequent CMAPs are evoked every 10 seconds, the decrement recovers to baseline in 1 to 2 minutes in myotonic dystrophy and myotonia congenita. Numbers on the left refer to the time in seconds following the exercise. In paramyotonia congenita, the recovery may be delayed, in the range of 10 to 60 minutes. (From McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10; reprinted by permission of John Wiley & Sons, Inc.)
exercise test may produce a drop in CMAP amplitude. The drop shows a marked delay in recovery to the baseline CMAP amplitude with repeated recording of the CMAP up to an hour if the test is performed when the muscle is cooled [13]. The short exercise test produces no decrement in the periodic paralysis syndromes. In contrast, in hyperkalemic and hypokalemic periodic paralysis, the prolonged exercise test often produces an immediate increase in the CMAP amplitude, especially if the initial amplitude is low. This increase is followed by a progressive drop in the CMAP amplitude by about 50% over 20 to 40 minutes, with most of the decline occurring in the first 20 minutes following prolonged exercise (Fig. 11) [11,12]. Summary Repetitive nerve stimulation and exercise testing are useful in the evaluation of patients with suspected disorders of the NMJ and muscle membrane excitability when performed with close attention to technical factors. They can be very helpful in the diagnosis of myasthenia gravis, Lambert-Eaton
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
205
Fig. 11. Typical pattern of response on prolonged exercise test in periodic paralysis. Following 3 to 5 minutes of prolonged exercise, the CMAP amplitude recorded every 1 to 2 minutes shows little change in normal controls (A). In the periodic paralysis syndromes (B), there is frequently an increment immediately after exercise, followed by a slow decrement over the next 20 to 30 minutes. Decrements greater than 40% are definitely abnormal. (From McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10; reprinted by permission of John Wiley & Sons, Inc.)
myasthenic syndrome, and botulism, as well as rare disorders of skeletal muscle membrane excitability, including paramyotonia congenita, myotonia congenita, myotonic dystrophy, and the periodic paralyses. References [1] Keesey JC. AAEM minimonograph no. 33: electrodiagnostic approach to defects of neuromuscular transmission. Muscle Nerve 1989;12(8):613–26.
206
B.E. Shapiro, D.C. Preston / Phys Med Rehabil Clin N Am 14 (2003) 185–206
[2] Kimura J. Electrodiagnosis in diseases of nerve and muscle. Philadelphia: FA Davis; 1989. [3] Oh SJ. Botulism: electrophysiological studies. Ann Neurol 1977;1(5):481–5. [4] Oh SJ. Diverse electrophysiological spectrum of the Lambert-Eaton myasthenic syndrome. Muscle Nerve 1989;12(6):464–9. [5] Preston DC, Shapiro BE. Repetitive nerve stimulation. In: Preston DC, Shapiro BE, editors. Electromyography and neuromuscular disorders. Boston: Butterworth-Heinemann; 1998. [6] Ptacek LJ, Johnson KJ, Griggs RC. Genetics and physiology of the myotonic muscle disorders. N Engl J Med 1993;18:482–9. [7] Shapiro BE, Ruff RL. Disorders of skeletal muscle membrane excitability: myotonia congenita, paramyotonia congenita, periodic paralysis and related syndromes. In: Katirji B, Kaminski H, Preston DC, Ruff RL, Shapiro BE, editors. Neuromuscular disorders in clinical practice. Woburn (MA): Butterworth-Heinemann; 2002. [8] AAEM Professional Practice Committee, Chiou-Tan FY, Gilchrist JM, Tim RW. Literature review of the usefulness of repetitive nerve stimulation and single fiber EMG in the electrodiagnostic evaluation of patients with suspected myasthenia gravis or Lambert-Eaton myasthenic syndrome. Muscle Nerve 2001;24:1239–47. [9] Jablecki C. AAEM case report no. 3: myasthenia gravis. Muscle Nerve 1991;14(5):391–7. [10] Tim RW, Sanders DB. Repetitive nerve stimulation studies in the Lambert-Eaton myasthenic syndrome. Muscle Nerve 1994;17(9):995–1001. [11] Kuntzer T, Flocard F, Vial C, Kohler A. Exercise test in muscle channelopathies and other muscle disorders. Muscle Nerve 2000;23:1089–94. [12] McManis PG, Lambert EH, Daube JR. The exercise test in periodic paralysis. Muscle Nerve 1986;9:704–10. [13] Streib EW. AAEE minimononograph no. 27: differential diagnosis of myotonic syndromes. Muscle Nerve 1987;10:603–15. [14] Hubbard JI. Microphysiology of vertebrate neuromuscular transmission. Physiol Rev 1973;53:674–723. [15] Slater CR, Lyons PR, Walls TH, Fawcett PRW, Young C. Structure and function of neuromuscular junctions in the vastus lateralis of man. Brain 1992;11:337–48. [16] Fatt P, Katz B. Spontaneous subthreshold activity at motor nerve endings. J Physiol 1952;117:109–28. [17] Paton WD, Waud DR. The margin of safety of neuromuscular transmission. J Physiol 1967;191:59–90. [18] Engel AG. Myasthenia gravis and myasthenic disorders. New York: Oxford University Press; 1999. [19] Lindstrom JM. Acetylcholine receptors and myasthenia. Muscle Nerve 2000;23:453–77. [20] Engel AG. Lambert-Eaton myasthenic syndrome. Ann Neurol 1987;22:193–9. [21] Lambert EH, Elmqvist D. Quantal components of end-plate potentials in the myasthenic syndrome. Ann N Y Acad Sci 1971;183:183–99. [22] Jablecki CK. Electrodiagnostic evaluation of patients with myasthenia gravis and related disorders. Neurol Clin 1985;3(3):557–72. [23] Tim RW, Massey JM, Sanders DB. Lambert-Eaton myasthenic syndrome: electrodiagnostic findings and response to treatment. Neurology 2000;54:2176–8. [24] Jablecki C. Lambert-Eaton myasthenic syndrome. Muscle Nerve 1984;7:250–7. [25] Schady W, MacDermott N. On the choice of muscle in the electrophysiological assessment of myasthenia gravis. Electromyogr Clin Neurophysiol 1992;32(3):99–102. [26] Schumm F, Stohr M. Accessory nerve stimulation in the assessment of myasthenia gravis. Muscle Nerve 1984;7(2):147–51. [27] Bernstein LP, Antel JP. Motor neuron disease: decremental responses to repetitive nerve stimulation. Neurology 1981;31(2):204–7. [28] Aminoff MJ, Layzer RB, Satya-Murti S, Faden AI. The declining electrical response of muscle to repetitive nerve stimulation in myotonia. Neurology 1977;27:812–6.