Check Your Diagnosis — Patient 26
Luciana DeSaibro, M.D.
Multifocal Motor Neuropathy with Conduction Block
Patient #26 presented with slowly progressive asymmetrical limb weakness, atrophy, and fasciculations without evidence of sensory or upper motor neuron involvement. The pattern of weakness was primarily distal, and cranial nerves were not affected. These findings indicate a lower motor neuron syndrome or asymmetrical motor neuropathy. The differential diagnosis includes the various causes of motor neuron disease, chronic inflammatory demyelinating neuropathy, motor neuropathies, and mononeuritis multiplex.
Motor neuronopathies (motor neuron degeneration) include amyotrophic lateral sclerosis, progressive muscular atrophies, adult-onset hexoseaminidase deficiency, segmental muscular atrophies, poliomyelitis, and post-polio syndrome. Despite his original diagnosis, ALS is not a strong clinical consideration in this patient as there is no evidence of upper motor neuron involvement. The duration of the disease process and lack of proximal involvement also argue against ALS as a diagnostic consideration. One of the spinal muscular atrophies could be considered, but the patient's age argues against even late onset cases. Spinobulbar muscular atrophy (Kennedy's syndrome) was not strongly considered as the patient had no evidence of bulbar involvement and did not have gynecomastia or testicular atrophy. Poliomyelitis causes acute, not chronic progressive motor neuron loss; there was no history of previous polio exposure to suggest a post-polio syndrome. Adult onset hexosaminidase deficiency causes an upper and lower motor neuron syndrome, not consistent with the clinical presentation in this case.
CIDP usually causes a symmetrical, distal, sensorimotor neuropathy primarily affecting the legs (though arms may be prominently affected, and proximal involvement is not unusual). Sensory symptoms are extremely common, and sensory signs are almost universal. However, rare cases of motor involvement without significant sensory signs have been described. In most cases, however, the motor involvement is symmetrical at onset and throughout the course of the disease. The clinical course is progressive with episodes of relapse. Reflexes are universally decreased and often absent. These features argue against CIDP as the cause of neuropathy in this case. The EMG/NCV findings also argue against CIDP, as motor conduction velocities are markedly slowed even in areas without conduction blocks in CIDP—findings not present in this case.
Mononeuritis multiplex (multiple mononeuropathies) may be caused by diabetes mellitus, vasculitis, sarcoidosis, infectious causes (HIV, Lyme, leprosy), HNPP (hereditary neuropathy with predisposition to pressure palsies), multiple nerve tumors (neurofibromatosis), and perineuromas. Clinically, mononeuritis multiplex typically presents with asymmetrical motor and sensory involvement and variable involvement of reflexes. It would be highly unusual for mononeuritis multiplex to present as a pure motor syndrome, as in this case. The long course of the disease without other accompanying features would also be highly unusual for any of the causes of multiple mononeuropathies.
Motor neuropathies, such as those caused by lead intoxication and porphyria, do not present with such an indolent course and are not asymmetrical. Porphyric neuropathy may affect upper extremities and cranial nerves but is frequently accompanied by confusion, convulsions, and abdominal pain. Its presentation is acute or subacute. Patients with lead neuropathy may also present with prominent upper limb involvement, but usually also show anemia, nausea, vomiting, and epigastric pain.
Hereditary neuropathies that may present similar to the patient in this case include HSMN, type 2 and some subtypes of proximal and distal motor neuropathies. However, the onset of symptoms is generally earlier than in this case, and the weakness is generally symmetrical. Legs are usually affected long before the arms, and there is generally a family history of neuropathy. The reflexes are also invariably absent or markedly reduced.
The clinical presentation in this case was felt to be most consistent with multifocal motor neuropathy with conduction block. This entity, closely related to CIDP, is a presumed autoimmune neuropathy associated with anti-GM1 antibodies and presenting with asymmetrical motor involvement with minimal sensory findings. The reflexes may be preserved (or slightly elevated) at the onset of the disease, causing the frequent misidentification with ALS. In this patient, anti-GM1 antibodies were found to be markedly elevated. EMG/NCV confirmed the diagnosis by demonstrating normal conduction velocities in areas not affected by conduction blocks (thus differentiating this entity from CIDP, which has delayed conductions in virtually all nerves, and hereditary neuropathies). Laboratory evaluation for the various causes of mononeuritis multiplex showed no other identifiable cause for this patient's neuropathy.
The patient was treated with intravenous human immunoglobulin (HIG) and has steadily improved over the past year.
Chronic neuropathy with multifocal, persistent motor conduction block was first described in 1982 by Lewis et al. Several authors subsequently drew attention to electrophysiologically similar multifocal neuropathy with conduction block and noted clinical similarity to motor neuron disease. Pestronk et al. reported, in 1988, two patients with a treatable form of multifocal motor neuropathy and noted elevated antibodies to GM1 ganglioside.
Multifocal motor neuropathy (MMN) is characterized by slowly progressive asymmetrical weakness and muscle atrophy, which may be accompanied by cramps, fasciculations, and myokymia. Early in the disease course weakness is more pronounced than atrophy in affected muscles. This is an important clinical feature distinguishing MMN from MND, in which atrophy is generally consistent with the degree of weakness. As MMN progresses, however, atrophy becomes more prominent and is usually present in some muscles at the time of diagnosis. The disease usually begins and remains more prominent in the upper extremities. The most striking clinical feature is the multifocal distribution of the weakness, which, in the beginning, may be localized within the territory of individual peripheral nerves. The affected nerves are not palpably enlarged, as in patients with hereditary neuropathies.
Although MMN has been characterized as a pure motor neuropathy, minor sensory symptoms are invariably present. At onset, reflexes are frequently preserved, but as the disease progresses, the reflexes are decreased and often absent. MMN occurs most frequently in young adults, runs an indolent course, and patients generally remain ambulatory. Occasional subacute courses over several months have been described. The clinical features of weakness, amyotrophy, and fasciculations with preserved reflexes cause confusion with amyotrophic lateral sclerosis. However, cranial nerve involvement is extremely rare in patients with MMN, and patients do not show upper motor neuron involvement as in ALS.
Conduction block in motor nerves is considered the electrodiagnostic hallmark of MMN. Motor conduction block (MCB) is the failure of a nerve action potential to propagate across a segment of intact myelinated nerve fiber. MCB leads to reduction of the amplitude and the area of a compound muscle action potential (CMAP) after stimulation of the nerve proximal to the affected segment. The CMAP is normal distal to the affected segment. In MMN motor conduction block occurs most frequently in the ulnar and median nerves, both proximally and distally. MCB is, however, not specific for MMN, as it occurs also in acute and chronic compressive neuropathies, Guillain-Barre syndrome, and CIDP.
Conduction block may be seen at any level of the peripheral nerve including the root, plexus, compression points (elbow for ulnar nerve, tibial head for peroneal nerve, etc.), or along a random segment. Outside the sites of MCB nerve conductions are normal or nearly normal. A unique electrophysiologic feature of MMN is that sensory conductions are typically normal, even across a segment of MCB. This helps to differentiate MCB in MMN from other entities, which typically show blocks to both motor and sensory conductions. The EMG in MMN typically shows evidence of axonal degeneration with fibrillation potentials and positive sharp waves in atrophic muscles.
The mechanism underlying MCB in MMN is not entirely clear. MCB has been attributed to demyelination of one or several internodes. It may involve blocking of Na+ and K+ channels at the nodal or paranodal regions by an autoimmune mechanism. Anti-GM1 antibodies may have their major effect (if any) at the nodes of Ranvier. Takigawa et al showed that anti-GM1 antibodies (in vitro) decrease sodium conductance and increase membrane leakiness through a mechanism requiring the presence of complement. In addition, in the absence of complement, potassium conductance is increased. It is unknown to what extent these changes in Na+ and K+ conductance are important in vivo.
Sural nerve biopsies from patients with MMN have ranged from normal to showing evidence of perivascular inflammation, demyelination, or reduced myelinated fiber density. Most of the studied patients had no sensory symptoms. The significance of these sensory abnormalities in a predominantly motor neuropathy is unclear. These findings suggest, however, that sensory nerves may be involved in MMN. Motor nerves at the sites of MCB show focal enlargement, nerve edema, demyelination, and onion bulb formation.
The pathogenesis of MMN is unclear, but there is substantial evidence suggesting an autoimmune mechanism. As previously mentioned, histological studies have shown inflammatory demyelination, antibodies to nerve gangliosides are generally elevated, and patients demonstrate a favorable response to immunosuppressive therapy. The role of anti-GM1 antibodies, which are elevated in approximately 80% of patients with MMN, is unclear. While they may be pathogenic, causing both interference with Na+ and K+ conductances and regions of demyelination, other antibodies may be involved as well. GM1 is abundant in mammalian nerve cell membranes, particularly at nodes of Ranvier and particularly in motor fibers. Santoro et al demonstrated the presence of anti-GM1 IgM at nodes of Ranvier in patients with MMN. Intraneural injection of serum from patients who are anti-GM1 antibody + with MMN produced electrophysiological findings suggestive of MCB and demyelination in rats. These findings suggest, but do not yet prove that anti-GM1 antibodies are the cause of MMN. At this point, these antibodies are considered a good marker for the presence of this disease. However, anti-GM1 antibodies are elevated in many other conditions, including Guillain-Barre syndrome, CIDP, and ALS. Consequently, the presence of anti-GM1 antibodies cannot be used as the sole means of diagnosing MMN.
Several immunosuppressive treatments have been attempted for patients with MMN, including oral steroids, intravenous steroids, cyclophosphamide, plasma exchange, and human immunoglobulin (HIG). Prolonged use of oral steroids has repeatedly been shown to be of no use, and plasma exchange is of limited benefit for patients with MMN. Intravenous cyclophosphamide followed by oral maintenance has been shown to result in functional improvement and reduction in anti-GM1 antibody titers. However, the potential adverse effects of Cytoxan limit its use.
Human immunoglobulin (HIG) is the treatment of choice for MMN based on the results of two independent trials. Improvement ranged from mild to complete. The onset of improvement occurred within 14 days, and the effects lasted an average of two months. Patients, therefore require frequent boosters to maintain their functional gains. Clinical improvement was not associated with a consistent decrease of anti-GM1 antibody titers and is not invariably accompanied by a reduction of MCBs. Nobile-Orazio et al. reported that the addition of oral Cytoxan permitted a progressive reduction in the frequency of maintenance HIG infusions.
The mechanism by which HIG improves the clinical symptoms and signs of MMN is unclear, but the rapid response to treatment suggests a physiological rather than anatomical mechanism. It is possible that HIG plays some role in interfering with Na+ and K+ channel blockade, but may also play a role in reversing areas of demyelination.
Treatment with HIG is easy, safe, and generally well tolerated. HIG is initially infused at a dose of 2 grams/kg in divided doses of 0.4 grams/kg on five consecutive days. Objective strength testing should be performed before infusion and two weeks after the initial course to obtain a measure of the therapeutic response. If patients respond, monthly boosters are given. If there is no improvement within one month of the initial dose, the diagnosis should be reassessed; a second course of treatment is generally tried at this point. Cytoxan is reserved for severe cases that do not respond well to HIG. Risks of HIG treatment include allergic reactions (especially patients with IgA deficiency - IgA levels must be obtained before starting HIG), hypotension, fever, rash, headaches, flu-like symptoms. More serious side effects include volume overload, congestive heart failure, aseptic meningitis, acute renal failure (especially in diabetics) and neutropenia.