Chapter 49. Myasthenia Gravis and Related Disorders of the Neuromuscular Junction

Included under this title is a group of diseases affecting the neuromuscular junction, the most important of which is myasthenia gravis. Most of these disorders exhibit the characteristic and striking features of fluctuating weakness and fatigability of muscle. Some degree of weakness is usually present at all times but it is made worse by activity. The weakness and fatigability reflect physiologic abnormalities of the neuromuscular junction that are demonstrated by clinical signs and by special electrophysiologic testing. As an aid to understanding the diseases discussed in this chapter, the reader should consult the discussion of the structure and function of the neuromuscular synapse given in Chap. 45.

The cardinal feature of myasthenia gravis, usually referred to simply as myasthenia, is fluctuating weakness of voluntary (skeletal) muscles, particularly those innervated by motor nuclei of the brainstem, i.e., ocular, masticatory, facial, deglutitional, and lingual. Manifest weakening during continued activity, quick restoration of power with rest, and dramatic improvement in strength following the administration of anticholinesterase drugs such as neostigmine are the other notable characteristics. Myasthenia is an immune disease in which circulating antibodies against components of the motor postsynaptic membrane and subsequent structural changes in that membrane explain virtually all the features of the disease.

Historical Note

Several students of medical history affirm that Willis, in 1672, gave an account of a disease that could be none other than myasthenia gravis. Others give credit to Wilks (1877) for the first description and for having noted that the medulla was free of disease, in distinction to other types of bulbar paralyses. The first reasonably complete accounts were those of Erb (1878), who characterized the disease as a bulbar palsy without an anatomic lesion, and of Goldflam (1893); for many years thereafter, the disorder was referred to as the Erb-Goldflam syndrome. Jolly (1895) was the first to use the name myasthenia gravis, to which he added the term pseudoparalytica to indicate the lack of structural changes at autopsy. Also it was Jolly who demonstrated that myasthenic weakness could be reproduced in affected patients by repeated faradic stimulation of the motor nerve and that the "fatigued" muscle would still respond to direct galvanic stimulation of its membrane. Interestingly, he suggested the use of physostigmine as a form of treatment but there the matter rested until Reman, in 1932, and Walker, in 1934, demonstrated the therapeutic value of the drug. The relationship between myasthenia gravis and tumors of the thymus gland was first noted by Laquer and Weigert in 1901, and in 1949, Castleman and Norris gave the first detailed account of the pathologic changes in the gland.

In 1905, Buzzard published a careful clinicopathologic analysis of the disease, commenting on both the thymic abnormalities and the infiltrations of lymphocytes (called lymphorrhages) in muscle. In 1973 and subsequently, the autoimmune nature of myasthenia gravis was established through a series of investigations by Patrick and Lindstrom, Fambrough, Lennon, and A.G. Engel (1977) and their colleagues (see further on). These and other references to the historical features of the disease can be found in the reviews by Viets and by Kakulas and Adams; A.G. Engel's monograph (1999) is an excellent comprehensive reference.

Clinical Manifestations

Myasthenia gravis, as the name implies, is a muscular weakness formerly with a grave prognosis. As mentioned, repeated or persistent activity of a muscle group exhausts contractile power, leading to a progressive paresis, and rest restores strength, at least partially. These are the identifying attributes of the disease and their demonstration, assuming that the patient cooperates fully, is usually enough to establish the diagnosis.

The special vulnerability of the neuromuscular junctions in certain muscles gives myasthenia a highly characteristic clinical appearance. Usually the eyelids and the muscles of eye movement, and somewhat less often, of the face, jaws, throat, and neck, are the first to be affected. Infrequently the initial complaint is referable to the limbs or to breathing. Specifically, weakness of the levator palpebrae or extraocular muscles is the initial manifestation of the disease in about half the cases, and these muscles are involved eventually in more than 90 percent of cases. Ocular palsies and ptosis are usually accompanied by weakness of eye closure, a combination that is virtually always indicative of the disease although it may be observed in certain muscular dystrophies. Diplopia is common in myasthenia, but it does not correspond to the innervatory pattern of a nerve; instead, it is the result of asymmetrical weakness of several muscles in both eyes. As the disease advances, it spreads insidiously from the cranial to the limb and axial muscles, but there are instances of fairly rapid development, sometimes initiated by an infection, usually respiratory. In rare cases, the distal extremity muscles may be involved, such as the "myasthenic hand" described by Janssen and colleagues. Symptoms may first appear during pregnancy or, more commonly, during puerperium or in response to drugs used during anesthesia.

In addition to certain circulating autoantibodies, inflammatory thymic abnormalities of several types are closely connected with the disease, as elaborated further on, and weakness may begin months or years before or after removal of a thymoma.

Particular ocular signs are highly characteristic of myasthenia. For example, sustained upgaze for 30 or more seconds will usually induce or exaggerate ptosis and may uncover myasthenic ocular motor weakness. Cogan described a twitching of the upper eyelid that appears a moment after the patient moves the eyes from a downward to the primary position ("lid-twitch" sign). Or, after sustained upward gaze, one or more twitches may be observed upon closure of the eyelids or during horizontal movements of the eyes. Repeated ocular versions when tracking a target or by an optokinetic stimulus may disclose progressive paresis of the muscles that carry out these movements. Unilateral painless ptosis without either ophthalmoplegia or pupillary abnormality in an adult will most often prove to be a result of myasthenia. Usually, there is subtle ptosis of the other eye that can be revealed by manually elevating the more affected eyelid. Attempts by the patient to overcome ptosis may impart a staring expression of the opposite eye. Bright sunlight is said to aggravate the ocular signs and cold to improve them. The application of an ice pack over the eye often relieves the ptosis for a brief period.

Muscles of facial expression, mastication, swallowing, and speech are affected in 80 percent of patients at some time in the illness, and in 5 to 10 percent, these are the first or only muscles to be involved. Less frequent is early involvement of the flexors and extensors of the neck, muscles of the shoulder girdle, and flexors of the hips. (This pattern may be associated with a special autoantibody as discussed later.) Of the trunk muscles, the erector spinae are the most frequently affected. In the most advanced cases, all muscles are weakened, including the diaphragmatic, abdominal, and intercostal, and even the external (skeletal muscle) sphincters of the bladder and bowel. As the disease progresses, the involvement of any group of muscles closely parallels their degree of weakness early in the disease. The clinical rule also holds that the proximal muscles are far more vulnerable than distal ones, as they are in most other forms of myopathy.

Another characteristic and understandable feature of myasthenic weakness is its tendency to increase as the day wears on or with repeated use of an affected muscle group but curiously, patients seldom volunteer this information. A few patients report paradoxical worsening on awakening, especially if they have not taken medication during the night. In general terms, therefore, myasthenia gravis may be conceived as a fluctuating and fatigable oculofaciobulbar palsy.

Other features conform to the topography and fatigability of the disease. The natural smile becomes transformed into a snarl; the jaw may sag, so that it must be propped up by the patient's hand; chewing tough food may be difficult and the patient may have to terminate a meal because of inability to masticate and swallow. It may be more difficult to eat after talking, and the voice fades and becomes nasal after sustained conversation. Women may complain of inability to fix their hair or makeup because of fatigue of the shoulders, or of difficulty in applying lipstick because they are unable to purse and roll their lips. Weakness of the neck muscles causes fatigue in holding up the head. In cases with generalized weakness, there is difficulty in retaining flatus because of weakness of the external rectal sphincter.

A peculiarity of myasthenic muscle contraction that may be observed occasionally is a sudden lapse of sustained posture or interruption of movement resulting in a kind of irregular tremor, similar to that of normal muscle nearing the point of exhaustion. A dynamometer demonstrates the rapidly waning power of contraction of a series of hand grips, and repetitive stimulation of a motor nerve at slow rates while recording muscle action potentials shows the same decremental strength in a quantitative fashion (see Fig. 45-4A and further on).

Weakened muscles in myasthenia gravis undergo atrophy to only a minimal degree or not at all. Tendon reflexes are seldom altered. Even repeated tapping of a tendon does not usually tax muscles to the point where contraction fails. Smooth and cardiac muscles are not involved and other neural functions are preserved. Weakened muscles, especially those of the eyes and back of the neck, may ache, but pain is seldom an important complaint. Paresthesias of the face, hands, and thighs are reported infrequently but are not accompanied by demonstrable sensory loss. The tongue may display one central and two lateral longitudinal furrows (trident tongue), as pointed out originally by Buzzard; the tongue may be atrophic in the MuSK (muscle-specific tyrosine kinase) form of disease (see further on).

Certain epidemiologic features of the disease are of clinical interest. Its prevalence is variously estimated to be from 43 to 84 per million persons and the annual incidence rate is approximately 1 per 300,000. The disease may begin at any age, but onset in the first decade is relatively rare (only 10 percent of cases begin in children younger than 10 years of age). The peak age of first symptoms is between 20 and 30 years in women and between 50 and 60 years in men. Under the age of 40, females are affected two to three times as often as males whereas in later life, the incidence in males is higher (3:2). Of patients with thymomas, the majority is older (50 to 60 years) and males predominate.

Familial occurrence of myasthenia is known, but it is rare. Many such cases prove to have one of the genetically determined myasthenic syndromes and not the acquired autoimmune form of disease (see further on). More common is a family history of one of the autoimmune diseases enumerated earlier. For example, in the series reported by Kerzin-Storrar and associates 30 percent had a maternal relative with a connective tissue disease, suggesting that myasthenia gravis patients inherit a susceptibility to autoimmune disease. Two of our patients have sisters with lupus. There have also been reports of the concurrence of myasthenia and multiple sclerosis, but this association is less certain. There is an increased representation of HLA-B8 and -DR3 haplotypes, as occurs in other autoimmune diseases, which is discussed further on.

Clinical Grading

To facilitate clinical staging of therapy and prognosis, the classification introduced by Osserman remains useful; it can be found in his monograph cited in the references and in previous editions of this book. This system has been replaced by a scheme suggested by a task force of the Myasthenia Gravis foundation (see Jaretzki et al) as reproduced here.

• Class I Any ocular muscle weakness
  • May have weakness of eye closure
  • All other muscle strength is normal
• Class II Mild weakness affecting other than ocular muscles
  • May also have ocular muscle weakness of any severity
• IIa Predominantly affecting limb, axial muscles, or both
  • May also have lesser involvement of oropharyngeal muscles
• IIb Predominantly affecting oropharyngeal muscles, respiratory muscles, or both
  • May also have lesser or equal involvement of limb, axial muscles, or both
• Class III Moderate weakness affecting other than ocular muscles
  • May also have ocular muscle weakness of any severity
• IIIa Predominantly affecting limb, axial muscles, or both
  • May also have lesser involvement of oropharyngeal muscles
• IIIb Predominantly affecting oropharyngeal muscles, respiratory muscles, or both
  • May also have lesser or equal involvement of limb, axial muscles, or both
• Class IV Severe weakness affecting other than ocular muscles
  • May also have ocular muscle weakness of any severity
• IVa Predominantly affecting limb and/or axial muscles
  • May also have lesser involvement of oropharyngeal muscles
• IVb Predominantly affecting oropharyngeal muscles, respiratory muscles, or both
  • May also have lesser or equal involvement of limb, axial muscles, or both
• Class V Intubation, with or without mechanical ventilation, except when employed during routine postoperative management. The use of a feeding tube without intubation places the patient in class IVb.

Others, for example, Compston and colleagues, have proposed a classification based on a constellation of the age of onset, presence or absence of thymoma, antibody level against acetylcholine receptor (AChR), and association with human leukocyte antigen (HLA) haplotypes. Their system is as follows: (1) myasthenia gravis with thymoma—no sex or HLA association, high AChR antibody titer; (2) onset before age 40, no thymoma—female preponderance and an increased association with HLA A1, B8, and DRW3 antigens; (3) onset after age 40, no thymoma—male preponderance, increased association with HLA A3, B7, and DRW2 antigens, low AChR antibody titer. The last group includes a proportion of older men with purely ocular symptoms (formerly Osserman type I). Classifications such as these are meant to capture certain types and contexts of myasthenia more than to convey the severity of illness.

Course and Prognosis

The course of the illness is extremely variable. Rapid spread from one muscle group to another occurs in some, but in others the disease remains unchanged for years before progressing or there is no progression. Remissions may take place without explanation, usually in the first years of illness, but these happen in less than half the cases and seldom last longer than a month or two. If the disease remits for a year or longer and then recurs, it then tends to be steadily progressive. Relapse may also be occasioned by the same events that in some cases preceded the onset of the illness, especially infections.

In Simpson's opinion, and this coincides with our observations, the danger of death from generalized myasthenia gravis is greatest in the first year after onset of the disease. A second period of danger in progressive cases is from 4 to 7 years after onset. After this time, the disease tends to stabilize and the risk of severe relapse diminishes. Fatalities relate mainly to the respiratory complications of pneumonia and aspiration. The mortality rate in the first years of illness, formerly in excess of 30 percent, is now less than 5 percent and with appropriate therapy virtually all patients lead productive lives.

An aspect of interest is the timing and frequency of conversion from ocular and restricted oropharyngeal patterns of weakness to more widespread involvement including the diaphragm. Bever and coworkers have confirmed the general impression that an increasing duration of purely ocular myasthenia is associated with a decreasing risk of late generalization of weakness. In a retrospective study of 108 patients, these authors found that only 15 percent of the observed instances of generalization occurred more than 2 years after isolated ocular manifestations.

A later age at onset was also associated with a higher incidence of fatal respiratory crises. In general, patients whose disease begins at a younger age run a more benign course. Grob and colleagues, who recorded the course of an astonishing 1,036 patients for a mean duration of 12 years, found that the clinical manifestations remained confined to the extraocular muscles and orbicularis oculi in 16 percent. Their data further indicated that localized ocular myasthenia present for only a month was associated with a 60 percent likelihood that the disease would generalize, but in those cases that remained restricted for more than a year, only 16 percent became generalized. In contrast, of 37 consecutive cases carefully studied by Weinberg with only ocular signs, 17 had more widespread weakness within a period of 6 years. Also informative in Grob's series was that in 67 percent the disease attained its maximum severity within a year of onset, and in 83 percent, within 3 years. It has been stated that the progression of symptoms is more rapid in male than in female patients.

It is not widely recognized that isolated muscle groups may occasionally remain permanently weak even when the ocular and generalized weakness has resolved. The muscles most often affected in this way are the anterior tibialis, triceps, and portions of the face.

The long-term outlook for children with myasthenia is better than it is for adults, and their life expectancy is only slightly reduced. Rodriguez and colleagues followed a group of 149 children for an average of 17 years; 85 of them had thymectomies, one of the main treatments for myasthenia as discussed further on. Approximately 30 percent of the nonthymectomized and 40 percent of the thymectomized patients underwent remission and were free of symptoms, usually in the first 3 years of illness. Those children with bulbar symptoms and no ocular or generalized weakness had the most favorable outcome.

Thymic and Systemic Disorders Associated with Myasthenia

A nonneoplastic lymphofollicular hyperplasia of the thymic medulla occurs in 65 percent or more of cases of myasthenia and thymic tumors occur in 10 to 15 percent. Thymomas with malignant characteristics may spread locally in the mediastinum and to regional lymph nodes but they rarely metastasize beyond these structures; when they do, the lungs and liver are usually affected. It should be emphasized that thymic enlargement and tumors may be missed in plain films of the chest and should be sought by CT scanning.

A striking degree of hyperplasia of the medulla of the thymus characterized by lymphoid follicles with active germinal centers is found in the majority of cases. Hyperplasia is even more frequent in younger patients in the third and fourth decades. The cells in the centers of the follicles are histiocytes surrounded by helper T lymphocytes, B lymphocytes, and plasma cells; immunoglobulin G (IgG) is elaborated in the germinal follicles. These resemble the cellular reaction observed in the thyroid tissue of Hashimoto thyroiditis. Because the latter disease has been reproduced in animals by injecting extracts of thyroid with Freund adjuvants, it had long ago been suggested that the so-called thymitis of myasthenia gravis is the result of a similar autoimmune sensitization but the inciting events for this process are entirely unknown. Immunosuppression with steroids causes involution of the thymus.

With regard to thymic tumors, two forms have been described: one composed of histiocytic cells like the reticulum cells in the center of the follicles, and the other predominantly lymphocytic and considered to be lymphosarcomatous. Some of the tumors have a high proportion of spindle-shaped cells. Overlapping types have been common. Thymic tumors may be unattended by myasthenia, though myasthenia has eventually developed in all of the cases under our observation, sometimes 15 to 20 years after the tumor was removed surgically. According to Bril and colleagues, the severity of myasthenic symptoms is no different in patients with thymoma than it is from that in patients without a tumor, but our impression has been that patients with tumors, particularly children, often have a peculiar clinical course. For example, we have observed unexpected sudden remissions and severe relapses, as well as resistance to medications.

Many contemporary studies, including more than 40 autopsies at our hospitals, have confirmed Erb's original contention that myasthenia gravis is a disease without a central nervous system lesion. The brain and spinal cord are normal unless damaged by hypoxia and hypotension from cardiorespiratory failure. Furthermore, the muscle fibers are generally intact, although in fatal cases with extensive paralysis, isolated fibers of esophageal, diaphragmatic, and eye muscles may undergo segmental necrosis with variable regeneration (Russell). Scattered aggregates of lymphocytes (lymphorrhages) are also observed, as originally noted by Buzzard, but none of these changes in muscle explains the widespread and severe weakness.

The main ultrastructural alterations occur in the motor endplate. These changes, elegantly demonstrated by A.G. Engel and associates (1976, 1977, 1987), consist of a reduction and simplification in the surface area of the postsynaptic membrane (sparse, shallow, abnormally wide, or absent secondary synaptic clefts) and a widening of the synaptic cleft (Fig. 49-1). The number and size of the presynaptic vesicles and their quanta of acetylcholine (ACh) are normal. The observation of regenerating axons near the junction, the many simplified junctions, and the absence of nerve terminals supplying some postsynaptic regions suggested to Engel and coworkers (1976, 1977, 1987) that there was an active process of degeneration and repair of the neuromuscular junction, particularly of the postsynaptic side.

Although not directly relevant to myasthenia, it is of interest that a number of curious neurologic disorders occur in association with thymoma. Among our own patients were 2 with "limbic encephalitis" with memory loss and confusion that could not be differentiated from the paraneoplastic variety of encephalitis (see Chap. 31), 1 case of midbrain encephalitis, 1 of Morvan's fibrillary chorea (discussed in Chap. 50), and 1 of aplastic anemia. Some of these neurologic processes are associated with antibodies directed against voltage-gated potassium channels (VGKC). Such cases appear in the literature, and all are considered to have a humoral immune basis.

Of biologic and even greater clinical importance is the coexistence of myasthenia gravis and other autoimmune diseases. Thyrotoxicosis with periodic paralysis (5 percent of myasthenic patients; see further on and Chap. 50), lupus erythematosus, rheumatoid arthritis, Sjögren syndrome, mixed connective tissue disease, anticardiolipin antibody, and (curiously) polymyositis have all been associated with myasthenia more often than can be explained by chance. A proportion of young women with myasthenia have moderately elevated titers of antinuclear antibody without the clinical manifestations of systemic lupus.

Etiology and Pathogenesis

The clear demonstration of an immunologic mechanism operative at the neuromuscular junction was the most significant development in our understanding of myasthenia gravis. Patrick and Lindstrom discovered that repeated immunization of rabbits with AChR protein obtained from the electric eel caused a muscular weakness (contrary to what is stated in some books, their discovery was not accidental). Lennon and colleagues recognized this model as being similar to that of myasthenia gravis. Soon thereafter, Fambrough and coworkers demonstrated that the basic defect in myasthenia gravis was a marked reduction in the number of ACh receptors on the postsynaptic membrane of the neuromuscular junction. These observations were followed by the creation of an experimental model of the disease and the demonstration that experimentally induced myasthenia had clinical, pharmacologic, and electrophysiologic properties identical with those of human myasthenia gravis (Engel et al, 1976). It was also shown that humoral antibodies directed against protein components of AChR could transfer the myasthenic weakness to normal animals, and that the weakness as well as the physiologic abnormalities could be reversed by the administration of anticholinesterase drugs. Thus, the accumulated evidence satisfied the criteria for the diagnosis of an autoantibody-mediated disorder (Drachman, 1990).

The present view is that myasthenic weakness and fatigue are a result of the failure of effective neuromuscular transmission on the postsynaptic side. The greatly reduced number of receptors and the competitive activity of anti-AChR antibodies (see later) produce postsynaptic potentials of insufficient amplitude to discharge some muscle fibers. Blocked transmission at many endplates results in a reduction in the contractile power of the muscle. This deficiency is reflected first in the ocular and cranial muscles that are both the most continuously active and have the fewest AChRs per motor unit. Fatigue is understandable as the result of the normal decline in the amount of ACh released with each successive impulse.

Antibodies to AChR protein are present in more than 85 percent of patients with generalized myasthenia and in 60 percent of those with ocular myasthenia (Newsom-Davis). The presence of receptor antibodies has proved to be a reasonably sensitive and reliable test of the disease, as discussed later. The manner in which the antibodies that are directed against proteins in the intracellular compartment (such as anti-MuSK discussed later) causes weakness is not known.

Neuromuscular transmission is therefore impaired in several ways: (1) the antibodies block the binding of ACh to the AChR; (2) serum IgG from myasthenic patients has been shown to induce an increase in the degradation rate of AChR. This may be the result of the capacity of antibodies to cross-link the receptors; (3) antibodies cause a complement-mediated destruction of the postsynaptic folds (Engel and Arahata).

Although the evidence that an autoimmune mechanism is responsible for the functional disorder of muscle in myasthenia gravis is incontrovertible, the source of the autoimmune response has not been established. Because most patients with myasthenia have thymic abnormalities and a salutary response to thymectomy, it is logical to implicate the lymphoid reaction in this gland in the pathogenesis of the disease. Both T and B cells from the myasthenic thymus are particularly responsive to the AChR, more so than analogous cells from peripheral blood. Moreover, the thymus contains "myoid" cells (resembling striated muscle) that bear surface AChR. It is not known with certainty that thymic myoid cells are the source of immunologic stimulation in myasthenia gravis. The most obvious objection is that such cells are even more abundant in the normal than in the myasthenic thymus (according to Schluep et al). Another suggested pathogenesis, yet unconfirmed, is that a virus with a tropism for thymic cells might alter such cells and induce antibody formation. A viral infection might at the same time have a potential for oncogenesis, accounting for thymic tumors, but this is all speculative. Scadding and associates have suggested a different mode of thymic involvement; they have shown that thymic lymphocytes from patients with myasthenia gravis can synthesize anti-AChR antibody, both in culture and spontaneously.

Diagnosis

In patients who present with a changeable, specifically fatigable, diplopia or ptosis and the typical myasthenic facies—unequally drooping eyelids, relatively immobile mouth turned down at the corners, a smile that looks more like a snarl, a hanging jaw supported by the hand—the diagnosis can hardly be overlooked. However, only a few patients display this fully developed syndrome. Ptosis, diplopia, difficulty in speaking or swallowing, or weakness of the limbs is at first mild and inconstant and may be mistaken for a cerebrovascular disease. However, the finding that sustained activity of small cranial muscles results in weakness (e.g., increasing droop of eyelids while looking at the ceiling or diplopia when fixating in lateral or vertical gaze or reading for 2 to 3 min) and that contraction improves after a brief rest is virtually diagnostic, even in the early stages of the disease. Any other affected group of muscles may be tested in similar fashion. The characteristic ocular signs have already been described. For confirmation, the measurement of specific antibody (anti-AChR), electromyography, and certain pharmacologic tests described below are necessary. Several special clinical problems and associated conditions are summarized further on.

Electrophysiologic Testing

Characteristic of myasthenia is a rapid reduction in the amplitude of compound muscle action potentials during a series of repetitive stimulations of a peripheral nerve at a rate of 3 per second (decrementing response as shown in Fig. 45-4A). Reversal of this response by neostigmine or edrophonium has been a reliable confirmatory finding in most cases. A decremental response to stimulation can usually be obtained most often from the proximal limb muscles followed by the facial and, to a lesser extent, the hand muscles, which may or may not be clinically weak. During a progressive phase of the disease or during corticosteroid therapy, a slight initial incrementing response may be obtained, not to be confused with the marked incrementing response after voluntary contraction that characterizes the Lambert-Eaton syndrome (see further on).

Single-fiber electromyography (EMG) represents an even more sensitive method of detecting the defect in neuromuscular transmission. This technique demonstrates an inconstancy of the normally invariant interval between the firing of muscle fibers connected to the same motor unit ("jitter"—see "Single-Fiber Electromyography" in Chap. 45) or complete blocking of successive discharges from single muscle fibers belonging to the same motor unit. The test requires a great deal of cooperation from the patient and that contraction of a muscle be sustained at just the right amplitude in order to isolate single muscle fibers from the same unit. It is also possible to detect such pairs of fibers by electrical stimulation of a nerve. Nerve conduction velocities and distal motor latencies are normal unless there is a coincident polyneuropathy.

Neostigmine Test

Almost as valuable as electrophysiologic testing is testing with the anticholinesterase inhibitors neostigmine and in the past, edrophonium a more rapidly acting agent. These drugs prolong and exaggerate the effects of ACh in the synapse and thereby produce an increment in muscle power in the patient with myasthenia. Edrophonium is not easily available in the United States at the time of this writing but neostigmine affords a longer time for observation, as noted in the next paragraph. The tests are performed in the following manner. After the estimation of strength in a cranial (usually the levator palpebrae or an extraocular muscle) or limb muscle (by dynamometry), or vital capacity, neostigmine is injected intramuscularly in a dose of 1.5 mg. Atropine sulfate (0.8 mg) should be given several minutes in advance to counteract the unpleasant muscarinic effects of neostigmine (salivation, sweating bronchorrhea, borborygmi, bowel cramps and, sometimes, diarrhea). Neostigmine may alternatively be given intravenously in a dose of 0.5 mg, but its effect is often too brief to be as useful. After intramuscular injection of neostigmine, objective improvement occurs within 10 to 15 min, reaches its peak at 20 min, and lasts up to 1 h, allowing for careful verification of the neurologic improvement. Many neurologists perform this test twice, once with an injection of saline as a control.

Alternatively, 1 mg (0.1 mL) of edrophonium is given intravenously; if this dose is tolerated and no definite improvement in strength occurs after 45 s, another 4 to 9 mg is injected. A total dose of 10 mg is rarely necessary. Most patients who respond do so after 3 to 5 mg has been administered. The mild muscarinic effects of edrophonium are blocked by pretreatment with atropine 0.8 mg subcutaneously as for neostigmine. The clinical effect of improved ptosis, extraocular movements, oropharyngeal function, arm and shoulder abduction, or vital capacity persists for no more than 5 min with edrophonium and 60 min with neostigmine.

One caution: with either drug, some patients deteriorate immediately, but briefly, as a result of an increase in pulmonary secretions. A positive test consists of visible (objective) improvement in muscle contractility, fusion of diplopia, or resolution of fatigable ptosis. Dynamometry and measurement of forced vital capacity serve as more objective markers of improvement, or lack of effect. The report of subjective improvement alone is not dependable and one must be distrustful of equivocal test results, which may occur with ocular palsies due to tumors, thyroid disease, Guillain-Barré syndrome (GBS), progressive supranuclear palsy, or carotid aneurysms (pseudoocular myasthenia).

A negative test with an anticholinesterase agent does not entirely exclude myasthenia gravis but is a strong point against the diagnosis. In a small number of patients with periodic and purely ocular symptoms who later prove to have myasthenia gravis, the edrophonium and neostigmine tests (and electrophysiologic studies and AChR antibody measurements) may be entirely normal during the first or even after several acute episodes. Only later, for inexplicable reasons, do these tests become positive. Finally, the anticholinesterase-inhibiting drugs carry a small risk of inducing ventricular fibrillation and cardiac arrest so that testing should be carried out where emergency support is accessible.

Measurement of Receptor Antibodies in Blood

The detection of anti-AChR antibodies provides a reasonably sensitive and highly specific test for the diagnosis of myasthenia. The radioimmunoassay method of detection is accurate and widely used. Serum antibodies are found in 80 to 90 percent of patients with generalized myasthenia gravis and in approximately 60 percent of those whose symptoms are restricted to the ocular muscles (Vincent and Newsom-Davis). For the most part, adults with myasthenia whose sera are persistently negative for AChR antibodies do not differ clinically or electromyographically from those with antibodies with the exception noted below. Persistently negative AChR antibody tests are more frequently found in patients with ocular myasthenia than in patients with generalized weakness. Patients with a thymoma and severe generalized myasthenia are practically always seropositive. Interestingly, the antibody titers usually remain elevated during clinical remissions.

Instances of "seronegative" disease are sometimes due to antibody production against unusual muscle epitopes that are located on or near the acetylcholine receptor; their detection requires a special panel of tests. However, the majority of such cases have been ascribed to IgG antibodies directed against an intracellular muscle-specific kinase (MuSK). This enzyme plays a role in supporting the normal structure of the postsynaptic membrane and in the arrangement of AChR but its main function may be in developmental synaptic differentiation. Scuderi and colleagues and others have proposed that patients with MuSK antibody, mostly women, have a special clinical syndrome of prominent oculobulbar weakness, often with severe disease and respiratory crises (see also Evoli et al). Others have reported a different pattern of mainly neck and proximal weakness that simulates a typical myopathy. Many of these patients are inadequately responsive to anticholinesterase treatment. Also of interest, but not currently used in routine diagnosis, is the presence of antibodies directed against striated muscle in almost half of myasthenic patients and an even higher incidence (stated to be 85 percent) in patients who also have a thymoma.

Each of the commonly used diagnostic tests, electrophysiology, edrophonium, and antibodies, proves to be about equally reliable. Kelly and coworkers obtained positive results with single-muscle-fiber recording in 79 percent, with the antireceptor antibody test in 71 percent, and with the edrophonium test in 81 percent. Combined, they confirmed the diagnosis in 95 percent of clinically suspected cases. Presumably, had the anti-MuSK receptor antibody test been available, the sensitivity of serologic diagnosis would have been higher.

In keeping with the observation of some myasthenic patients that their weakness improves in the cold, a test has been devised in which an ice pack is placed over a ptotic eyelid for 2 min or to the limit of the patient's tolerance. Sethi and colleagues found that ptosis was diminished in 8 of 10 patients. In our patients, this effect has not been as consistently evident, but it may be a useful adjunctive test.

Other diagnostic tests performed routinely in essentially all patients with myasthenia gravis include CT of the chest (for the detection of thymic enlargement or thymoma), tests of thyroid function for reasons discussed further on, and in cases of uncertain diagnosis, magnetic resonance imaging of the cranium and orbits to exclude compressive and inflammatory lesions of the cranial nerves and ocular muscles.

Special Diagnostic Problems

We have encountered the following clinical problems in myasthenia:

1. The concurrence of myasthenia gravis and thyrotoxicosis. Thyrotoxicosis may produce a characteristic ocular myopathy and there is a tentative relation to periodic paralysis as indicated in Chap. 50. There is no certain evidence that thyrotoxicosis aggravates myasthenia gravis; some have even observed an inverse relationship between the severity of the two conditions. Hypothyroidism, however, does worsen the myasthenic symptoms. The ophthalmoplegia of thyrotoxicosis can usually be distinguished by the presence of an associated exophthalmos (early in the disease, exophthalmos may be absent), lack of ptosis, and the lack of definitive response to neostigmine. Polymyositis and inclusion body myopathy are differentiated from myasthenia by lack of involvement of extraocular muscles, but they may affect oropharyngeal muscles, as does myasthenia. Finding the signs of these diseases in combination with those of myasthenia indicates a concurrence of two independent autoimmune diseases.

2. The neurasthenic or depressed patient who complains of weakness when actually referring to fatigability. There is no ptosis, strabismus, or dysphagia, though an anxious individual may complain of diplopia (usually of momentary duration when drowsy) and also of tightness in the throat (globus hystericus). A number of such patients claim improvement with neostigmine but objective and reversal is always uncertain. Conversely, myasthenia is as often mistaken for hysteria or other emotional illness, mainly because the physician is unfamiliar with myasthenia (or with hysteria) and has been overly impressed with the precipitation of the illness by an emotional crisis. Furthermore, fatigability is a feature of all of these conditions, but only in the psychiatric ones does it extend to the sphere of mental endurance. Those with myasthenia do not usually complain of fatigue of the mind, whereas these are frequent complaints in psychiatric conditions. A similar problem arises frequently on our services in judging breathlessness due to anxiety or cardiopulmonary disease in a patient with presumed myasthenia. Careful appraisal of the breathing pattern and determination of the vital capacity or other spirometric measurements are helpful here.

3. Progressive external ophthalmoplegia and other restricted myopathies, including the congenital myasthenic states. These may be mistaken for long-established myasthenia gravis. It should be emphasized that the extraocular muscles and levator palpebrae may be permanently damaged by myasthenia and cease to respond to neostigmine. Another possibility is that restricted ocular myasthenia may not respond to anticholinesterase drugs from the beginning and the diagnosis of myasthenia is erroneously excluded. One must then turn to other muscles for clinical and electromyographic and serologic confirmation of the diagnosis.

4. Myasthenia with dysarthria and dysphagia, but without ptosis or obvious strabismus. These may be confused with multiple sclerosis, polymyositis, inclusion body myopathy, stroke, motor neuron disease, or some other neurologic disease. Testing with an anticholinesterase inhibitor, single-fiber and repetitive stimulation recording, and measurement of antibodies usually clarifies the matter.

5. The initial manifestations of botulism—blurred vision, diplopia, ptosis, strabismus, and ophthalmoparesis—may be mistaken for myasthenia gravis of acute onset. In botulism, however, the pupils are usually large and unreactive, and the eye signs are followed in rapid succession by involvement of bulbar, trunk, and limb muscles.

   Similarly, the oculopharyngeal-brachial and variants of Guillain-Barré syndrome (GBS) in the early stages have many of the features of myasthenia, including ptosis, that may be partially responsive to anticholinesterase drugs. The loss of tendon reflexes, acral paresthesias and areflexia, or the development of ataxia in the limbs make the diagnosis of GBS at once apparent and detailed electrophysiologic testing distinguishes the two conditions.

6. Intoxication with organophosphate insecticides, because of their capacity to induce a cholinergic crisis, may be confused with a myasthenic crisis (see further on).

Certain other small clinical points may be helpful in differentiating myasthenia from other diseases that affect the cranial musculature. A hanging jaw and hanging head are indicative of myasthenia, whereas complete or asymmetric facial paresis is typical of GBS. Botulism usually affects the pupillary convergence reaction, and GBS does so only when there is complete internal and external ophthalmoplegia; diphtheria affects mainly the accommodative reaction early on. The question of midbrain stroke as a consequence of basilar artery occlusion arises in a case with total ophthalmoplegia; it should be recalled that the level of consciousness is usually reduced if vertical gaze and pupillary reactions are lost in cases of basilar artery stroke; such is not the case in neuromuscular diseases. The myasthenic syndrome of Lambert-Eaton, discussed further on, only occasionally affects the ocular muscles, but is identified by its other clinical and electrophysiologic features. Ocular paresis, as may occur in nemaline polymyopathy, oculopharyngeal dystrophy, and thyrotoxic ophthalmic disease, come on too slowly in most cases to be confused with myasthenia gravis. On occasion, the eye movements in myasthenia simulate an internuclear ophthalmoplegia or other "central" sign, even to the extent of including nystagmus in an abducting eye.

Treatment

The treatment of this disease involves the careful use of two groups of drugs—anticholinesterases and immunosuppressants including corticosteroids and in special acute circumstances, plasma exchange and intravenous immune globulin. An elective thymectomy is appropriate in many patients as discussed below.

Anticholinesterase Drugs

The two drugs that give the best results in ameliorating myasthenic weakness are neostigmine (Prostigmin) and pyridostigmine (Mestinon), the latter being preferred by most clinicians and patients. The usual dose of pyridostigmine is 30 to 90 mg given every 6 h (typically a 60-mg pill is tried first); the oral dose of neostigmine ranges from 7.5 to 45 mg given every 2 to 6 h. Extended-action forms of both drugs are available but are given at bedtime mainly to patients who complain of weakness during the night or early morning hours. The dosage of these drugs and their frequency of administration vary considerably from patient to patient, but we agree with Drachman (2003) that the maximal useful dosage of pyridostigmine rarely exceeds 120 mg given every 3 h. Table 49-1 lists the approximate dose-equivalents of these various drugs.

For mild cases, for patients in partial remission after thymectomy, and for purely ocular myasthenia, the use of anticholinesterase drugs may be the only form of therapy necessary for some period of time (ocular myasthenia often responds well to small doses of corticosteroids as noted further on). Although these drugs seldom relieve symptoms completely (the response of ocular symptoms is typically incomplete), most such patients are able to function well.

Corticosteroids

For the patient with moderate to severe generalized weakness who is responding inadequately to anticholinesterase drugs, the long-term administration of corticosteroids is the most consistently effective form of treatment, as described in a large series of patients by Pascuzzi and colleagues. Small doses of corticosteroids (prednisone 15 to 25 mg daily) alone or in combination with azathioprine (see later) are also often adequate to control ocular myasthenia. However, one must be prepared to contend with the side effects of long-term corticosteroid therapy and we hesitate to undertake such a program in children or patients with severe diabetes or other diseases that are likely to be aggravated. Because recent experience with the newer immunosuppressive agents was not incorporated into most prior series, the uniform use of steroids might not be correct.

The usual form of corticosteroid therapy is prednisone (or corresponding doses of prednisolone), beginning with 15 to 20 mg/d and increasing the dose gradually until a satisfactory clinical response is obtained or until a daily dose of 50 to 60 mg is reached. With higher doses or more rapid elevations of the doses, worsening of weakness in the first weeks may occur and hospitalization and careful observation for respiratory difficulty may be advisable. Improvement after the initiation of corticosteroids occurs over a few weeks. Once the maximal effect from prednisone has been attained, the dosage can be reduced gradually over months to the lowest point at which it is still effective. Our practice has been to then attempt to institute an alternate-day schedule, which diminishes the side effects; some patients have done better with a modest difference in dose from one day to the next, rather than omitting a dose entirely on alternate days. Potassium supplements and antacids should be prescribed liberally if needed, as with any chronic corticosteroid regime and consideration should be given to prophylaxis with antibiotics for Pneumocystis infection, and bisphosphonate for osteoporosis if long-term treatment is anticipated. At the outset of steroid therapy, anticholinesterase drugs are given simultaneously; as the patient improves, the dosage of the latter may be adjusted downward.

Azathioprine and Other Immunosuppressive Drugs

Azathioprine is a useful adjunct to steroids in patients who cannot tolerate or fail to respond to prednisone. It has been possible to manage the disease reasonably well in a few of patients with azathioprine alone, but there is no study to support this practice (see Palace et al, 1998). Treatment typically begins with 50 mg (1 tablet) bid for a few days; if this is tolerated, the dosage is raised to 2 to 3 mg/kg/d (150 to 250 mg daily). However, improvement occurs much more slowly than with corticosteroids and a significant response may not be evident for many months to a year (Witte et al). Liver function tests and blood cell count should be checked regularly. The Myasthenia Gravis Clinical Study Group found that the most severe forms of the disease, particularly those resistant to either prednisone or azathioprine alone, benefit from the combination of the two medications. Azathioprine is a prodrug of mercaptopurine, which is metabolized principally by thiopurine methyltransferase (TPMT). Approximately 3 per 100,000 persons are deficient in the enzyme, for which reason, some clinicians measure its level before initiating azathioprine in order to avoid bone marrow toxicity; it has not been our practice to do so. There are many variant alleles of TPMT and a larger number of patients have partial deficiency of the enzyme or even excessive enzyme activity but it has not been clear how to utilize this information in myasthenia. Azathioprine interacts with other drugs such as allopurinol and warfarin.

Cyclosporine is another immunosuppressive drug that has shown benefit in clinical trials (Tindall et al). It is given in 2 divided doses daily, to a total of 6 mg/kg, but not often used currently because of serious side effects (hypertension, nephrotoxicity) and its high cost. Because of the success of alternative regimens, we have had occasion over the years to use cyclosporine only once for myasthenia.

Mycophenolate is currently being used as an adjunct to corticosteroids and has been beneficial in several small trials but failed to demonstrate similar effect in larger controlled studies. The clinical improvement, when it does occur, has generally occurred sooner than it does with azathioprine (Meriggioli et al). Diarrhea was the main adverse effect. Several experts in the field believe that mycophenolate is preferable to most of the adjunctive medications and in some milder cases may be effective alone, but reconciling this view with recent failed trials is vexing.

De Feo and coworkers have used cyclophosphamide administered in intravenous pulses; they were able to remove 5 of their 12 patients from steroids, but the appropriate use of this potent agent is not clear and we have resorted to it infrequently. Drachman and colleagues (2003), as well as others, describe a regimen of high-dose cyclophosphamide (50 mg/kg/d for 4 consecutive days) followed by granulocyte-stimulating factor to "reboot" the immune system in refractory cases. This approach has risks but may be justified if all other measures have failed. Liver function and white blood cell count require monitoring. On the basis of pilot trials, many other drugs, for example, tacrolimus as reported by Ponseti and colleagues, rituximab, and etanercept have come into use in patients who are dependent on or resistant to corticosteroids, including those with antibodies to MuSK (Diaz-Manero et al).

Plasma Exchange and Intravenous Immune Globulin

For severe myasthenia that is refractory to treatment with anticholinesterase drugs and prednisone, or during an acute period of worsening, one must resort to other measures. Striking temporary remissions (2 to 8 weeks) may be obtained by the use of plasma exchange. This form of treatment may be lifesaving during a myasthenic crisis. It also finds use before and after thymectomy and at the start of immunosuppressive drug therapy. Plasma exchange is also helpful in limiting the aforementioned weakness that is often induced by the institution of high-dose corticosteroids. The number and volume of exchanges required in these circumstances is somewhat arbitrary but they tend to be less than those required for GBS; several exchanges of 2 to 3.5 L each (totaling approximately 125 mL/kg) performed over a week usually suffice. The removed plasma is replaced with albumin and saline. It has been estimated that a 2-L exchange will remove 80 percent of circulating antibodies and that this will be reflected in reduced ACh antibody levels in 3 to 5 days. There is only an approximate correlation between a reduction in the titer of anti-AChR antibody and the degree of clinical improvement. In a crisis requiring plasma exchanges and mechanical ventilation, it has been our practice to discontinue or curtail the use of anticholinesterase drugs and resume them as the patient is being weaned from the ventilator. Also, it may be that sensitivity to these drugs may be enhanced in the hours after an exchange so that their dosages must be adjusted accordingly.

A small number of patients respond so well to plasma exchange and find the side effects of steroids so intolerable that they choose to be maintained with two to three exchanges every several weeks or months. Immunoadsorption, a technique similar to plasma exchange that removes antibodies and immune complexes by passing blood over a tryptophan column, is less cumbersome than conventional plasma exchange and has been effective, but experience with this procedure is limited.

Intravenous immune globulin is similarly useful in the short-term control of acutely worsening myasthenia. The usual dose is 2 g/kg given in divided doses over 3 to 5 days. Several small series suggest that the effect is equivalent to a series of plasma exchanges. However, plasma exchange and immune globulin have been subjected to only limited systematic study or comparison and, while these treatments are invaluable in deteriorated patients or those in crisis, they offer only short-term benefit. In two separate small studies, Gajdos and colleagues and Barth et al found no difference between intravenous immune globulin (IVIg) and plasma exchange (1997) and no difference between IVIg 1 g/kg/d given for 1 or 2 days (2005) in myasthenic exacerbations, most cases of which were less severe than typical crisis.

Thymectomy

This operation, first introduced by Blalock, despite the absence of proof in trials, is considered an appropriate procedure for many patients with generalized myasthenia gravis between puberty and 55 years of age. The surgery is performed electively and not during an acute deterioration of myasthenia. The remission rate after thymectomy is approximately 35 percent provided that the procedure is done in the first year or two after onset of the disease, and another 50 percent will improve to some extent (Buckingham et al). The remission rate is progressively lower, but not negligible, if the operation is postponed beyond this time. In patients with myasthenia restricted to the ocular muscles for a year or longer, the prognosis is so good that thymectomy is unnecessary. The response to thymectomy is not evident for several months and is maximal in most cases by 3 years. In favorably responding cases, levels of circulating receptor antibody are reduced or disappear entirely. If possible, thymectomy should be postponed until puberty because of the importance of the gland in the development of the immune system, but juvenile myasthenia is also quite responsive. The results are not as predictable in patients who harbor a thymoma.

A suprasternal approach for removal of the gland has been developed and results in less postoperative pain and morbidity than occurs with a transsternal thoracotomy but the transsternal operation may be preferable because it assures a more complete removal of thymic tissue. Thymectomy is best performed in a hospital where there is close collaboration between the thoracic surgeon and the neurologist. If the patient is weak preoperatively, a course of plasma exchange or immune globulin may be given preceding the surgical procedure. Large "stress doses" of corticosteroids seem to be unnecessary in most patients who have been taking these medications chronically. After operation, respiratory assistance must be available if needed. Neostigmine intramuscularly, may be given every 3 to 6 h postoperatively. Usually the dose requirement is about 75 percent of that taken before surgery. As improvement occurs, oral medications are resumed as remission is not anticipated for many months or longer as noted above.

Thymectomy may also be a safe and effective treatment in elderly patients with myasthenia. In 12 such individuals, Olanow and associates reported complete remission in 9 and clinical improvement in the remainder. The improvement in older patients is less convincing than it is in the younger group, in part because the thymus is atrophic. Nonetheless, some of our patients who were older than age 60 years did benefit.

Removal of the thymus gland is also indicated in practically all patients in whom thymoma is detected by CT scanning of the chest. The tumor can be locally invasive but rarely metastasizes. The operative approach is through the anterior thorax, with adequate exposure to remove all the tumor tissue. If the tumor cannot be removed completely, the remaining tissue should be treated with focused radiation. Local spread and lymph node invasion have been treated with combinations of chemotherapy including cisplatin, but it is not very satisfactory. Park and colleagues concluded from a large retrospective study of metastatic cases that chemotherapy offers some benefit in terms of survival, but this remains controversial.

Despite this endorsement of thymectomy for generalized myasthenia, it has remained an unproven therapy by a modern trial and attempts to recruit patients for such an endeavor have been difficult.

Myasthenic and Cholinergic Crisis

A rapid and severe deterioration of the myasthenia itself, termed myasthenic crisis, can bring the patient to the brink of respiratory failure and quadriparesis in a matter of hours. A respiratory infection or excessive use of sedative medications or drugs with a potential for blocking neuromuscular transmission may precede the myasthenic crisis. We have encountered numerous cases in which oropharyngeal weakness has led to aspiration pneumonia, which, in turn, precipitated a crisis. Just as often, a precipitating event is not evident. Rarely, a respiratory arrest is the first manifestation of crisis. Such events may occur at any time after the diagnosis of myasthenia, but half are evident within 12 to 18 months. In an experience with 53 patients in myasthenic crisis at the Columbia-Presbyterian Medical Center, pneumonia was the most frequent precipitating event, but no cause could be determined in almost one-third of cases (Thomas et al).

Incipient respiratory failure is usually marked by a reduction of vital capacity, often accompanied by restlessness, anxiety, diaphoresis, or tremor. Once the diaphragm fails, movements of the chest wall and abdomen become paradoxical (the abdomen moves inward during inspiration) or there may be shallow excursions of the chest, alternating with paradoxical movements as discussed in Chap. 26 where the characteristics of neuromuscular respiratory failure are described. In an emergency, after clearing of the airway, such a patient can be supported briefly by a tight-fitting face mask and manual bag (Ambu) breathing. The chest wall will be very compliant as a result of muscular weakness.

Management of the crisis entails timely and careful intubation followed by mechanical ventilation in a critical care unit that is equipped to attend to the medical and neurologic needs of such patients. Respiratory failure in a few patients can be managed by the use of bilevel positive airway pressure (BIPAP) according to Rabinstein and Wijdicks, but in our experience, this has not been consistently effective in avoiding endotracheal intubation. One must cope with both the oropharyngeal weakness and secretions that endanger the airway, and the diaphragmatic weakness. Anticholinesterase drugs, which exaggerate secretions, are best withdrawn at the time of intubation. A useful maneuver is to allow the patient to remain off cholinergic drugs for several days while on a ventilator; there is often a heightened response to the reinstitution of medications after this period. The use of plasma exchange or intravenous gamma globulin as described earlier, is equivalently effective in hastening improvement and weaning from the ventilator. Some of our colleagues have used high-dose corticosteroid infusions in these circumstances but this measure has not been particularly successful in our unit and, in the short run, carries the risk of inducing worsening of the weakness (Panegyres et al).

Patients generally respond equivalently to plasma exchange or immunoglobulin infusions in 1 or 2 days, but more often a week or more is required for recovery after a full course of 4 to 5 exchanges or 3 to 5 g/kg IVIg given in divided daily doses. Whether the previously mentioned studies (e.g., Gajdos et al) comparing the two treatments and comparing doses of IVIg in myasthenic exacerbations are pertinent to crisis is unknown but we almost always institute one or the other soon after it as it is apparent that respiratory failure is imminent or worsening.

It is generally best to wait 2 or 3 weeks before committing a patient to tracheostomy. When weaning from the ventilator is anticipated, anticholinesterase agents are reintroduced slowly, and treatment with corticosteroids can be instituted if necessary. Oral doses of 60 mg pyridostigmine or 15 mg neostigmine are roughly equivalent to 0.5 to 1 mg neostigmine intravenously and 1.5 to 2 mg intramuscularly, as noted in Table 49-1. The management of the critically ill patient with myasthenia is reviewed in the monograph by Ropper and colleagues.

Most patients with myasthenic crisis take several weeks to recover, and a few of our patients have remained ventilator-dependent for months. In the extensive experience of 53 patients from Columbia-Presbyterian, half of the patients could be safely extubated within 2 weeks and three-quarters by a month (Thomas et al). There were 7 deaths among 53 patients, reflecting the gravity of this syndrome even in the modern era of intensive care. Atelectasis, severe anemia, congestive heart failure, and clostridial diarrhea (associated with antibiotic use) portend a prolonged period of generalized weakness and intubation. From time to time, one encounters a patient in whom respiration and ambulation do not improve for many months after a myasthenic crisis. In our experience these have been middle-aged or older patients, usually women, in whom an element of hyperthyroidism or hypothyroidism may have been operative. They become wasted as the proximal limb and axial muscles, including the diaphragm, fail to recover their power, even though the ocular and oropharyngeal muscles improve. The role of corticosteroids in producing a concomitant proximal myopathy is a consideration that can be solved by careful electrophysiologic examination.

If the response to anticholinesterase drugs is poor and progressively larger doses are not relieving symptoms, there is always the danger of a cholinergic crisis. In our own experience with more than 60 patients of severe myasthenia in an intensive care unit, we have been persuaded of the occurrence of a cholinergic crisis only rarely. This consists of a relatively rapid increase in muscular weakness, usually coupled with the adverse muscarinic effects of the anticholinesterase drug (nausea, vomiting, pallor, sweating, salivation, bronchorrhea, colic, diarrhea, miosis, bradycardia). An impending cholinergic effect is betrayed by constriction of pupils. If the blood pressure falls with bradycardia, 0.6 mg atropine sulfate should be given slowly by the intravenous route. Neostigmine or repetitive stimulation may be used to determine whether or not weakness is to the result of an excess of anticholinesterase medications. However, this test has been misleading and undoubtedly has contributed to an overestimation of the frequency and importance of the cholinergic crisis. Infection, or the natural course of the disease, has been far more common causes of acutely worsening weakness and respiratory failure.

The only recourse in cases of long-standing and severe myasthenia is to continue an average dose of corticosteroids, immunosuppressive, and anticholinesterase medications with intermittent trials of immune globulin or plasma exchanges. This is also a desperate situation in which high-dose cyclophosphamide followed by granulocyte-stimulating factor, as mentioned earlier, may result in slow improvement. Other agents such as rituximab may be tried.

Management of Anesthesia and Pregnancy in the Myasthenic Patient

These represent special problems. Surgical procedures of any type are often sufficiently stressful to produce decompensation of the disease. If the patient is unable to take medications orally, anticholinesterase agents may be given intramuscularly (approximately one-thirtieth of the oral dose of pyridostigmine and one-tenth the oral dose of neostigmine listed in Table 49-1). If corticosteroids were being used they may be continued and the dose generally left unchanged; large "stress" doses are generally unnecessary, as mentioned earlier in the discussion of thymectomy. Neuromuscular blocking agents of the noncompetitive type may have a very prolonged effect in these patients and should be avoided as part of the anesthetic regimen. If they are necessary for some reason, a period of mechanical ventilation should be anticipated. In contrast, the dose of succinylcholine (which is not recommended) required to produce muscle relaxation may be larger than usual. Any drug, the use of which is contemplated in anesthetic and postsurgical management, should be checked against the list of agents that are capable of exaggerating myasthenic weakness (see further on).

Pregnancy is usually uncomplicated in patients with myasthenia but some women who are partially treated for myasthenia and have generalized weakness may have difficulty in assisting with vaginal delivery. However, the use of intravenous cholinesterase inhibitors is contraindicated because of the possibility of inducing uterine contractions, and cytotoxic drugs are generally avoided during pregnancy because of the potential for fetal abnormalities. Also, magnesium is not recommended for the treatment of eclampsia because its neuromuscular blocking effects may worsen myasthenic weakness. Delivery usually proceeds normally, and breast-feeding is not thought to be a problem with regard to the transmission of AChR antibodies. Almost half of women with myasthenia have an exacerbation of varying degree in the several weeks postpartum. A rapidly dropping level of alpha-fetoprotein has been implicated as this protein inhibits binding of antiacetylcholine antibodies to postsynaptic receptors. The issues of neonatal myasthenia and of reduced intrauterine movements with arthrogryposis are considered later.

(See Table 49-2)

Considered here are several disorders of neuromuscular transmission characterized clinically by muscular weakness and fatigability but differing in mechanism from autoimmune myasthenia gravis. The Lambert-Eaton myasthenic syndrome, neonatal myasthenia, the congenital myasthenic syndromes, and the myasthenic syndromes induced by drugs and toxins are the main disorders in this group. Two additional important diseases—botulism and organophosphate poisoning—are described elsewhere in the book.

The Myasthenic-Myopathic Syndrome of Lambert-Eaton (Lambert-Eaton Syndrome)

This special form of myasthenia, observed most often in patients with oat cell carcinoma of the lung, was first described by Lambert, Eaton, and Rooke in 1956 and further by Eaton and Lambert in 1957. Unlike myasthenia gravis, the muscles of the trunk, shoulder girdle, pelvic girdle, and lower extremities are the ones that become weak and fatigable. The first symptoms are difficulty in arising from a chair, climbing stairs, and walking; the shoulder muscles are usually affected later. Although ptosis, diplopia, dysarthria, and dysphagia may occur, presentation with these symptoms is distinctly unusual. Increasing weakness after exertion stamps the condition as myasthenic, but in direct contrast to myasthenia gravis, there may be a temporary increase in muscle power during the first few contractions. The tendon reflexes are often diminished but complete abolition of the reflexes should raise the question of an associated carcinomatous polyneuropathy. Fasciculations are not seen.

One of the most instructive reviews of this disease is that by O'Neill and colleagues: In 50 well-studied cases they described proximal leg weakness in all, arm weakness in 39, diplopia in 25, ptosis in 21, and dysarthria in 12. Other complaints were paresthesias, aching pain (suggesting arthritis), and a number of autonomic disturbances, such as dryness of the mouth, constipation, difficult micturition, and impotence. This latter group of symptoms gives the syndrome an unmistakable stamp as discussed further on under "Diagnosis." One should not be surprised to find other neurologic manifestations of neoplasia (e.g., polyneuropathy, polymyositis or dermatomyositis, multifocal leukoencephalopathy, cerebellar degeneration, as discussed in Chap. 31).

The onset of weakness is subacute and the course, variably progressive. Males are affected far more often than females (5:1). The weakness may precede discovery of the tumor by months or years. Approximately 60 percent of cases are associated with small cell lung cancer, but small numbers have also occurred with carcinoma of the breast, prostate, stomach, and rectum, and with lymphomas; in about one-third of patients, no tumor is found. Some cases are associated with other autoimmune diseases, but most are paraneoplastic or idiopathic. The condition may occur in children, usually with no relation to tumor. In the tumor cases, death usually occurs in a few months or years from the effect of the neoplasm; the idiopathic ones fluctuate over years.

The response to neostigmine and pyridostigmine is poor or at least unpredictable, and this finding in a myasthenic patient should bring the diagnosis of Lambert-Eaton syndrome to mind. In contrast, d-tubocurarine, suxamethonium chloride, gallamine, and other muscle relaxants have a deleterious effect and may cause fatality, just as in myasthenia gravis.

Conventional electrodiagnostic studies show no abnormality in the peripheral nerves. A single stimulus of nerve may evoke a low-amplitude muscle action potential (in contrast to myasthenia gravis, in which it is normal or nearly so) whereas at fast rates of stimulation (50 per s as shown in Fig. 45-4B) or following strong voluntary contraction (for 15 s or longer) there is a marked increase in the amplitude of action potentials (incrementing response hence the term "inverse myasthenia," which has been applied to Lambert-Eaton syndrome [LEMS]). Single-fiber recordings show an increase in "jitter" as in myasthenia gravis, as described in Chap. 45.

Elmquist and Lambert, from a series of studies of excised muscle, deduced that there is a defect in the release of ACh quanta from the presynaptic nerve terminals, akin to the effects of botulinum toxin, magnesium excess, and neomycin. The presynaptic vesicles themselves appear to be normal in morphology and content. Also in contrast to myasthenia gravis, the surface area of the postsynaptic receptor membrane in this syndrome is actually increased (A.G. Engel, 1976).

The physiologic mechanism in Lambert-Eaton myasthenic syndrome is a loss of voltage-gated calcium channels on the presynaptic motor nerve terminal. The calcium channels become cross-linked and aggregated by IgG autoantibodies, ultimately reducing the number of functioning channels (Fukunaga et al). These antibodies against a specific component of the presynaptic membrane have the effect of reducing the presynaptic release of ACh, virtually the opposite of myasthenia gravis.

A serologic test for these antibodies is available and is performed to confirm the diagnosis. Even in patients without detectable antibodies against voltage-gated calcium channels, passive transfer experiments indicate the presence of a circulating factor with similar activity. Muscle biopsy is normal or shows only the same slight, nonspecific changes as in myasthenia gravis. The thymus is, of course, normal.

Recognition of the Lambert-Eaton syndrome should lead to a search for an occult tumor, particularly of the lung. A positron emission tomography (PET) scan of the body may be useful for this purpose, although CT of the lungs is usually adequate. If found, it should be treated; this alone may result in improvement in the neurologic syndrome. If none is found, the search should be repeated at regular intervals, as the tumors at first are small and may be inapparent even at autopsy. Numerous cases of the more typical paraneoplastic syndromes discussed in Chap. 31, such as cerebellar degeneration, may coexist with the Lambert-Eaton syndrome, most of them are a result of small cell lung cancer (see Mason and colleagues) while others a result of idiopathic antibodies to the calcium channels.

Treatment

Most patients with this disease benefit from administration of 3,4-diaminopyridine (3,4-DAP), an agent that blocks potassium channels in the distal motor terminal, thus prolonging depolarizations and enhancing the release of ACh vesicles. The drug is given as 20 mg, up to five times a day, either alone or in conjunction with pyridostigmine (Lundh et al). It is not approved by the FDA in the United States and must be obtained from specialty pharmacies but nonetheless, has replaced the formerly used guanidine hydrochloride that had hematologic and renal toxicity with long-term use. With regard to long-term relief, numerous regimens have been tried and favored by different groups. Streib and Rothner were able to achieve improvement with prednisone. Dau and Denys have claimed the best results in nontumor cases with repeated courses of plasmapheresis in combination with prednisone and azathioprine. Intravenous immune globulin has also been effective in a few reported cases. Bain and coworkers indicate that the benefit is a result of reduction in calcium channel autoantibodies, but the mechanism whereby intravenous immune globulin produces this effect could not be established. Many clinicians prefer alternate-day administration of prednisone and azathioprine (prednisone 25 to 60 mg/d, and azathioprine 2 to 3 mg/kg body weight daily) supplemented intermittently as needed by intravenous immune globulin. The response to treatment tends to be slow, over a period of months and sometimes up to a year. Some patients recover fully; in others, restoration of power is incomplete.

Diagnosis

A syndrome of symmetrical weakness and fatigability of proximal muscles coupled with dry mouth, sphincter disturbances, aching muscles, and diminished reflexes are diagnostic. The illnesses with which it might be confused are myasthenia gravis, inclusion body myopathy, and polymyositis. There is a superficial resemblance to hysterical paralysis, where the patient may do better with encouragement on making a succession of voluntary contractions, and arthritis, where pain hampers the first movements more than the successive ones. Then the electrodiagnostic and specific serologic tests are of value.

Neonatal Myasthenia Gravis

An estimated 10 to 20 percent of babies born to mothers with myasthenia show transient signs of myasthenia (hypotonia, weak cry and suck). This transitory phenomenon is apparent at birth and has a mean duration of about 2 to 5 weeks; recovery is usually complete within 2 months of birth (rarely longer), without later relapse. Uncommonly, the mother with myasthenia reports reduced intrauterine movements, suggesting a dangerous degree of myasthenia in the fetus. A few of these children will be born with arthrogryposis, the result of a sustained period of intrauterine immobility, and this complication tends to recur in subsequent births.

It has long been assumed that neonatal myasthenia is a result of the passive transplacental transfer of AChR antibodies. This explanation is not entirely satisfactory insofar as maternal AChR antibodies are transferred from mother to fetus in all AChR antibody-positive pregnancies and the incidence and severity of neonatal myasthenia gravis do not correlate with the severity or duration of the mother's illness or with the serum level of the maternal AChR antibody. In fact, neonatal myasthenia may occur when the mother is in remission.

Administration of plasma exchange and anticholinesterase drugs to the infant may be useful in hastening recovery from neonatal myasthenia.

Congenital Myasthenic Syndromes (See Table 49-3)

Sporadically in the medical literature, there have appeared reports of a benign congenital myopathy in which myasthenic features could be recognized in the neonatal period or soon thereafter. The affected infants had been born to mothers who did not have myasthenia and were in the past described under headings such as "Myasthenia Gravis in the Newborn" and "Familial Infantile Myasthenia" (Greer and Schotland; Robertson et al) to distinguish the condition from passively transmitted neonatal myasthenia.

In the 1970s and 1980s, after the autoimmune basis of myasthenia gravis was established and its morphologic and physiologic features defined, the differences between this disease and the familial infantile forms became evident. Since then, at least eight distinct and rare congenital myasthenic syndromes have been delineated on the basis of their electrophysiologic and ultrastructural features, and a number of others have been partially characterized.

As indicated in Table 49-3, the congenital myasthenic syndromes are inherited defects in components of the presynaptic, synaptic, or postsynaptic junctional apparatus. Broadly speaking, the defects involve resynthesis or packaging of ACh or a paucity of synaptic vesicles (presynaptic); deficient quantal release; a deficiency of endplate ACh esterase (synaptic); or kinetic abnormalities in the AChR channels, or AChR deficiency (postsynaptic). It has been estimated that in three-fourths of the cases, the defect is postsynaptic.

These disorders are distinguished by neonatal onset, fluctuating and sometimes progressive weakness that may be quite severe, sometimes pronounced muscle hypotrophy, persistent ptosis, and seronegativity for anti-AChR and anti-MuSK antibodies. Moreover, heritability (typically autosomal recessive) is suggested by familial occurrence of the disorders among siblings. The most important clue to the disease in the neonate is an increase in ptosis and in bulbar and respiratory weakness with crying. Later in infancy these symptoms, as well as fluctuating ocular palsies and abnormal fatigability, are brought out by other types of sustained activities. Motor milestones may be delayed. In some cases, the myasthenic weakness and fatigability do not become evident until the second and third decades of life. Testing with anticholinesterase drugs is inconsistently positive in a few forms of congenital myasthenia as mentioned further on, but it usually is negative.

Two of the congenital myasthenic diseases—fast channel syndrome and slow channel syndrome—are consequences of mutations in AChR subunits that accelerate (fast channel) and slow (slow channel) the kinetics of gating of receptor channel (Croxen et al). Another well-characterized type, usually causing arthrogryposis and recurrent apneic spells but occasionally having an adult onset (as late as 48 years) has been traced to mutations in the "rapsyn" gene. The rapsyn protein plays a role in maintaining the integrity of the postsynaptic membrane (see Burke et al). Deficiency of the enzyme required to synthesize and package ACh in vesicles (choline acetyltransferase) causes a congenital myopathy with episodes of stress-induced apnea. In another type, the synaptic vesicles form poorly and are reduced in number. Being largely presynaptic defects, both of these disorders respond to acetylcholine esterase (AChE) inhibitors. In comparison, children with congenital myasthenia as a result of deficiency of AChE deteriorate markedly if given AChE inhibitors. Three disorders affecting postsynaptic structures (the fast-channel syndrome, AChR deficiency, and the myasthenic syndromes associated with deficiencies of rapsyn and plectin) also respond to AChE inhibition and 3,4-DAP, although both agents are hazardous in individuals with the slow-channel syndrome.

Another recently discovered congenital myasthenia is caused by a recessive mutation in DOK-7 that causes a simplified structure of the synapse but no alteration in acetylcholine receptors (Beeson et al). The clinical features as described by Palace and colleagues (2007) are of a limb-girdle pattern of weakness that causes a delay in walking after the child has reached other normal motor milestones and of ptosis from an early age.

A further category of congenital syndromes of defects in the presynaptic quantal release of acetylcholine have been described by Milone and colleagues and an intriguing prenatal myasthenic disease called Escobar syndrome (arthrogryposis, pterygia, and respiratory distress) has been traced to mutations in the gamma subunit of the acetylcholine receptor, a component expressed only in fetal life and replaced with maturity by the epsilon subunit (see Hoffman and coworkers).

The studies of A.G. Engel have systematically defined and classified these disorders in a series of extensive investigations of more than 100 cases. A detailed account of this work can be found in his review with Ohno and Sine and in the chapter on the subject in his monograph Myasthenia Gravis and Myasthenic Disorders.

Myasthenic Weakness Caused by Antibiotics and Other Drugs and by Natural Environmental Toxins (See Chap. 43 and Table 49-2)

Many drugs can cause a myasthenic syndrome or a worsening of myasthenia gravis by their action on pre- or postsynaptic structures. In the case of a nonmyasthenic patient, this is most likely to happen in the presence of hepatic or renal disease that allows excessive accumulation of the causative agent. The myasthenic state in these conditions is acute and lasts hours or days, with full recovery provided that the patient does not succumb to respiratory failure. The ocular, facial, and bulbar muscles are involved, just as in native myasthenia. The treatment is to provide respiratory support, discontinue the offending drug, and attempt to reverse the conduction block at the motor endplate by infusions of calcium gluconate, potassium, and anticholinesterases, along the lines suggested by Argov and Mastaglia.

There are more than 30 drugs in current clinical use (other than anesthetic agents) that may, under certain circumstances, interfere with neuromuscular transmission in otherwise normal individuals. Of these, the most important are the aminoglycoside and quinolone antibiotics. Myasthenic weakness has been reported with 18 different antibiotics but particularly neomycin, kanamycin (less so with gentamicin), colistin, streptomycin, polymyxin B, and certain tetracyclines (McQuillen et al; Pittinger et al). It has been shown that these drugs impair transmitter release by interfering with calciumion fluxes at nerve terminals. The fluorinated quinolones (fluoroquinolones)—typified by ciprofloxacin—affect both pre- and postsynaptic activity. They are especially hazardous when given to patients with myasthenia but they may be used if necessary to treat infections in patients who are already receiving ventilatory support.

Other agents—particularly the organophosphate insecticides and nerve gases—cause paralysis by binding to cholinesterase and blocking the hydrolysis of ACh. The endplate remains depolarized and is refractory to neural stimuli. The most notable of these agents are (1) botulinum toxin, which binds to cholinergic motor endings, blocking quantal release of ACh; (2) black widow spider venom, which causes a massive release of ACh, resulting in muscular contraction and then paralysis from a lack of ACh; (3) d-tubocurarine, which binds to AChR; (4) suxamethonium and decamethonium, which also bind to AChR; (5) organophosphates, which bind irreversibly to AChE; and (6) malathion and parathion, which inhibit AChE. The actions of all these agents except for the organophosphate "nerve gases" are transitory.

The administration of d-penicillamine has also caused an unusual type of myasthenia. The weakness is typical in that rest increases strength—as do neostigmine and edrophonium—and the electrophysiologic findings are also the same. In such cases, Vincent and associates (1978) found anti-AChR antibodies in the serum; hence, one must assume that this is a form of induced autoimmune myasthenia gravis. In these respects it differs from the weakness caused by aminoglycosides (see review by Swift). Rarely, typical autoimmune myasthenia gravis develops as part of a chronic graft-versus-host disease in long-term (2- to 3-year) survivors of allogeneic marrow transplants.

A large group of naturally occurring environmental neurotoxins are known to act at the neuromuscular junction and to induce muscle paralysis of a pattern like that of myasthenia gravis. Venoms of certain snakes, spiders, and ticks are common and well-known animal poisons, as are ciguatera and related toxins (from fish that have ingested certain dinoflagellates), curare (from plants), and Clostridium botulinum—all of which are discussed in other parts of this book, particularly Chap. 43. Poisoning by these natural neurotoxins constitutes an important public health hazard in many parts of the world. This class of disorders of neuromuscular transmission has been reviewed by Senanayake and Roman.