Repetitive stimulation

The classic electrophysiologic demonstration of an NMJ transmission defect is the documentation of a decremental response of the compound muscle action potential (CMAP) to repetitive stimulation (RS) of a motor nerve [Oh, 1988]. The decrement is due to failure of some muscle fibers to reach threshold and contract when successive volleys of ACh vesicles are released at the NMJ. Failure to reach the threshold EPP to achieve muscle contraction is called blocking. The percentage decrease in amplitude and area is calculated between the first CMAP produced by a train of stimuli and each successive one. In most laboratories, five or six responses are obtained at 2 or 3 Hz, and the maximal percentage decrement can be measured at the fourth or fifth response. A decrement of greater than 10 percent is considered a positive RS study (Figure 91-3).

Fig. 91-3 Repetitive stimulation of the ulnar nerve at 3 Hz, record adductor digiti minimi.

A, Baseline, abnormal amplitude decrement of 27 percent. B, Immediately after 10 seconds of exercise the decrement has resolved, demonstrating repair. NPamp, negative peak amplitude; NParea, negative peak area; Resp, response.

In some patients, a decremental response can be demonstrated at baseline. However, often a brief period of exercise (usually 1 minute) is required to fatigue the NMJ so that the decrement can be observed. This phenomenon of postexercise exhaustion (PEE) usually occurs at 2–4 minutes after exercise (Figure 91-4). In addition, repair or an improvement in the decrement can sometimes be observed immediately (within seconds) after brief exercise (see Figure 91-3).

Fig. 91-4 Repetitive stimulation of the ulnar nerve at 3 Hz, record adductor digiti minimi.

A, At baseline there is only a borderline decrement at response 4. B–E, After exercise, a 12–13 percent decrement develops immediately (B) and 1 minute (C) post exercise, and this worsens at 2 and 4 minutes (D and E), demonstrating post-exercise exhaustion. F, After 10 seconds of brief exercise, the decrement improves. NPamp, negative peak amplitude; NParea, negative peak area; Resp, response.

RS typically is first recorded in a distal thenar or hypothenar muscle after stimulating the median or ulnar nerve, respectively. If no decrement is observed, RS can be performed on a proximal limb muscle (i.e., trapezius, deltoid, biceps) or a facial muscle (orbicularis oculi). An arm board is used to immobilize the hand muscles. False-positive results are more of a problem in proximal limb muscles because of motion artifact.

Because RS is a reflection of the integrity of NMJ transmission, a decrement is more often observed in clinically weak muscles. Thus, even if a patient has generalized MG, a decrement in a hand muscle is unlikely if there is only facial and proximal limb weakness. In a pure ocular MG patient, a decrement may not be present in the orbicularis oculi, unless that muscle is weak on examination.

As with the edrophonium test, RS does not have to be performed on every MG patient if the diagnosis is certain based on clinical findings and a positive anti-AChR antibody.

A protocol for RS of the ulnar nerve recording over the adductor digiti minimi follows:

Apply electrodes and immobilize the hand.

Obtain a normal baseline CMAP and increase to supramaximal intensity.

If there is a small CMAP amplitude at baseline, screen for LEMS before carrying out RS (see later).

After establishing a stable baseline CMAP, give five shocks at 3 Hz.

If there is no decrement, exercise the hand by having the patient spread the fingers for 1 minute. Repeat RS immediately after exercise, and at 1, 2, 4, and 6 minutes post exercise.

If there is a decrement after exercise (PEE), briefly exercise the muscle again for 10 seconds and repeat RS at 3 Hz. If the decrement now improves, this indicates repair.

If there is a marked decrement at baseline to 3-Hz RS (>20 percent), exercise for 10 seconds only and repeat immediately. If the decrement improves, this is also repair. In this case, the protocol for demonstrating PEE described previously is optional.

A decremental response is more likely present in a proximal muscle than in a distal muscle. In the series by Stalberg and Sanders, a decrement in a distal muscle was reported in 38 percent of patients, whereas a decrement in proximal muscles occurred in 64 percent [Stalberg and Sanders, 1981]. Similar findings have been described by other authors [Vial et al., 1991; Oh et al., 1992]. In a study of 27 juvenile myasthenic patients, Afifi and Bell found that the chance of finding a decrement doubled to 66 percent by including proximal muscles. In children with generalized disease at onset, 80 percent showed a decrement when proximal muscles were studied. In ocular MG, decrements are less common, occurring in 20–50 percent of patients [Stalberg and Sanders, 1981; Evoli et al., 1988]. Facial muscle RS should be included when clinical suspicion for anti-MuSK myasthenia exists, as facial muscles are much more prominently involved in this group [Muppidi and Wolfe, 2009]. RS at faster rates (i.e., 20 or 50 Hz) usually is not performed unless there is concern about LEMS, a rare condition in children.

Single-fiber electromyography

Single-fiber electromyography (EMG) is a more sensitive measure of neuromuscular transmission than RS and can be considered in select children. In MG, the time required for the EPP at the NMJ to reach threshold is extremely variable. The measurement of this variability in the EPP rise time is known as jitter. The jitter value, calculated in microseconds, is the most important piece of data obtained from single-fiber EMG. Everyone, including healthy individuals, has some degree of jitter (Figure 91-5). Myasthenic patients have increased jitter values (Figure 91-6). In addition, blocking occurs in myasthenics if a muscle fiber’s EPP never reaches threshold and depolarization does not occur. The frequency of blocking, expressed as a percentage, is also determined with single-fiber EMG (see Figure 91-6). In healthy individuals, the percentage of blocking is 0 percent.

Fig. 91-5 Recruited single-fiber electromyography of the extensor digitorum communis muscle.

A, The single-fiber needle is inserted in the muscle, with the patient extending the digit so that action potentials from at least two muscle fibers from the same motor unit are recorded. A trigger is placed on the initial potential under the arrow. As both potentials are from the same motor unit, the second potential (under the arrowhead) will appear at approximately the same time in relation to the triggered potential. B, One hundred successive acquisitions are obtained. The degree of variability in the time that the second potential occurs in the successive pairs is the jitter, measured as the mean consecutive difference (MCD). In this normal pair, the MCD is 14.3 μsec (normal <35 μsec).

Fig. 91-6 Recruited single-fiber electromyography from the extensor digitorum communis in a patient with myasthenia gravis.

A, Three muscle fiber potentials from the same motor unit are recorded. B, One hundred successive recordings are captured. The arrow identifies the triggered potential. The degree of jitter for potential 1 is 270 μsec and for potential 2 is 187 μsec (normal <35 μsec). In addition, potential 1 demonstrates blocking in 50 percent of the acquisitions. (Two of these instances are indicated by the closed circles in B.)

Single-fiber EMG is undoubtedly the most sensitive test for MG in adults. It is abnormal in 94 percent of generalized and 80 percent of ocular MG patients [Sanders and Howard, 1986; Oh et al., 1992]. However, single-fiber EMG has several disadvantages. It is a tedious and lengthy study that requires considerable patient cooperation and is poorly tolerated by many children. The need to use nondisposable single-fiber electrodes was also a limitation, but recently, normative data for single-fiber studies with disposable concentric needles have been published [Stalberg and Sanders, 2009]. Stimulated single-fiber EMG can be performed under sedation, requires less patient cooperation, and may be preferred in children, although it is still a lengthy procedure [Jabre et al., 1989]. An abnormal single-fiber EMG study is not specific for MG because increased jitter commonly occurs as a result of other neuromuscular diseases, including motor neuron disease, peripheral neuropathy, and many myopathies [Oh, 1988]. Fortunately, it is seldom necessary to perform single-fiber EMG to diagnose MG in children. It is probably most useful in children who present difficult diagnostic dilemmas and who otherwise have normal laboratory studies for MG. Single-fiber EMG abnormalities may be seen in 12–33 percent of first-degree relatives of patients with juvenile MG [Stalberg et al., 1976; Anlar et al., 1995]. Conventional needle EMG has limited diagnostic value in MG; however, short-duration, small-amplitude, early recruited myopathic units can be seen in anti-MuSK MG patients [Padua et al., 2006].

Anti-AChR antibodies

Finding elevated AChR-antibody levels in the serum of a suspected MG patient is the most specific and reassuring diagnostic test. When the AChR-antibody assay is positive, it can be argued that no other diagnostic studies for MG are needed. Most hospitals and medical centers do not run the assay in house, but the test is commercially available through a number of reference laboratories. It generally takes a few days to receive the results from the reference laboratory. Thus, the clinician who is initially making the diagnosis is often in the position of performing the edrophonium and electrophysiologic tests while awaiting the serologic results.

Anti-AChR antibody levels are not elevated in all MG patients. The assay is most helpful in adult generalized MG; it is positive in 85 percent of such patients [Lindstrom et al., 1976; Vincent and Newsom-Davis, 1985; Oh et al., 1992; Drachman, 1994]. Ocular MG patients, however, have measurable anti-AChR antibodies in only 50 percent of cases [Provenzano et al., 2009]. Children represent another group of MG patients who are often antibody-negative. In one study, 50 percent of prepubertal children with autoimmune juvenile MG were seropositive [Andrews et al., 1994]. Seropositive rates of 68 and 91 percent were observed in peripubertal and postpubertal disease onset, respectively. Similarly, seropositivity was more common in girls with onset of juvenile MG after 11 years of age [Anlar et al., 1996]. Seronegativity was more common in pure ocular forms, mild disease, and remission [Afifi and Bell, 1993].

Because congenital myasthenic syndromes and seronegative autoimmune MG present in early childhood, differentiating these disorders when the family history is negative is often difficult [Andrews, 2004]. Fluctuating weakness or disease severity and good responses to immunotherapy favor an autoimmune basis [Anlar et al., 1996].

The most common anti-AChR antibody test is the binding radioimmunoassay using bungarotoxin, measured in nanomoles per liter. The upper limit of normal varies among reference laboratories (usually between 0.03 and 0.5 nmol/L) [Lennon, 1982]. Other assays that block bungarotoxin binding to AChR (“blocking” assay) or that reduce the density of AChR on cultured human myotubes (modulating antibody assay) are also commercially available [Howard et al., 1987]. These additional assays may be useful in patients with suspected MG who test negative with the standard binding assay [Howard et al., 1987], but do not add significantly to the diagnostic sensitivity. Some laboratories offer all three (binding, blocking, and modulating) antibodies as one serological test. Recently, low-affinity anti-AChR antibodies against rapsyn-clustered AChR were seen in 66 percent of patients who were otherwise seronegative. These were mainly IgG1 antibodies that can activate complement C3b deposition [Vincent et al., 2008]. These new low-affinity AChR assays are not yet commercially available.

Anti-AChR antibody titers correlate poorly with MG severity [Roses et al., 1981]. Although the titer often falls as the clinical condition improves, antibody titers in general do not guide therapeutic decisions. Indeed, MG patients in clinical remission may still have elevated titers, but this is not an indication to continue immunosuppressive therapy.

Anti-MuSK antibodies

Since 2001, IgG from 40–70 percent of seronegative generalized patients has been found to bind to the extracellular domain of muscle-specific receptor tyrosine kinase (MuSK) [Hoch et al., 2001; Sanders et al., 2003; McConville et al., 2004]. Marked female predominance with mean age of onset in the fourth decade has been typical [Evoli et al., 2003; Sanders et al., 2003], although Evoli et al. encountered a child with disease onset at age 6 years, and two patients from the initial series presented before age 10 [Hoch et al., 2001]. The earliest reported onset of anti-MuSK MG is 2 years [Murai et al., 2006]. Three main patterns of anti-MuSK MG have been observed; one of them is clinically indistinguishable from anti-AChR generalized MG. The other two patterns are severe oculobulbar weakness and prominent neck, shoulder, and respiratory involvement, largely sparing ocular musculature. In these two phenotypic variants, limb strength is relatively intact [Sanders et al., 2003; Muppidi and Wolfe, 2009]. Anti-MuSK antibodies rarely occur in pure ocular MG [Wolfe et al., 2007].

It has been hypothesized that anti-MuSK antibodies impede agrin-mediated clustering of AChR and disrupt normal postsynaptic architecture [Jha et al., 2006]. We consider testing for anti-MuSK antibodies in all suspected MG patients who are anti-AChR antibody-negative. The assay is commercially available. Anti-MuSK MG is somewhat more refractory to conventional treatment when compared to anti-AChR MG [Pasnoor et al., 2009].

Striated muscle antibodies and other laboratory studies

Antibodies to striated muscle in MG patients were discovered before anti-AChR antibodies. These antibodies can be directed against a number of muscle proteins, including myosin, actin, alpha-actinin, titin, and ryanodine (RyR). It is generally believed that, if anti-striated muscle antibodies are present in an MG patient, they should raise suspicion for thymoma, as they are reported in up to 84 percent of patients with thymoma [Limburg et al., 1983]. However, these antibodies may be found in patients without thymoma and in patients with thymoma who do not have MG [Cikes et al., 1988; Romi et al., 2000]. The absence of anti-striated antibodies also does not rule out thymoma. From an adult perspective, anti-titin and anti-RyR antibodies have been forwarded as a marker for more severe disease in MG patients presenting after age 40 [Romi et al., 2000]. Thyroid function tests are routinely obtained at the time of initial evaluation, as thyroid disease often coexists with MG [Meriggioli and Sanders, 2009].