Home, Search, Index, Links, Pathology, Molecules, Syndromes,
Muscle, NMJ, Nerve, Spinal, Ataxia, Antibody & Biopsy, Patient Info


Recent evidence shows that several pure motor neuropathy syndromes can be distinguished from amyotrophic lateral sclerosis (ALS). Identification of these motor neuropathy syndromes is important as, in contrast to ALS, they are often immune- mediated and treatable. The motor neuropathy syndromes usually have characteristic, but not unique, patterns of weakness and no upper motor neuron signs. Additional laboratory evaluation, including electrodiagnostic studies and measurement of serum autoantibodies, is usually needed to clearly distinguish these disorders from other demyelinating neuropathies and motor neuron syndromes.

Historical Aspects of Lower Motor Neuron Syndromes

The original descriptions of pure motor syndromes without upper motor neuron signs were probably cases of "progressive muscular atrophy" in the writings of Duchenne, Aran and others during the 19th century . Benign, focal motor neuron disorders, such as monomelic amyotrophy, were subsequently reported . These syndromes were usually considered as variants of ALS, as early pathological studies suggested that the primary focus of the disease was on cell bodies in the ventral horn.

A pathological report by Rowland et al. first documented that a patient with a pure motor syndrome could have the primary site of disease along the course of the axon. This patient, with a lower motor neuron (LMN) syndrome and a serum IgM M-protein, had damage to motor axons but not cell bodies. Motor neuropathies were first diagnosed during life by electrodiagnostic testing. Nerve conduction studies showed blockade of impulses at focal sites along the course of motor axons (motor conduction block) providing strong evidence that the primary site of disease lay in the peripheral nerve rather than the cell body. The phenomenon of conduction block had been described earlier in patients with sensory-motor neuropathies (chronic inflammatory demyelinating polyneuropathy (CIDP)) . Conduction block was thought to result from focal regions of immune-mediated demyelination along the course of the nerve.

In 1986 a patient was reported with a LMN syndrome without conduction block, but with a serum IgM M-protein that bound to GM1 ganglioside. In this instance the association of the motor syndrome with an autoantibody directed against a neural antigen suggested that the disorder might be immune-mediated. However, attempts at immunosuppression had no effect on the progressive disease in that patient. A clinical response to immunotherapy remains a "gold standard", without which it is difficult to argue that a syndrome is immune mediated. In 1988 two patients with a multifocal motor neuropathy, motor conduction block, and serum IgM anti-GM1 antibodies were reported to improve after treatment with cyclophosphamide . It now appears that either motor conduction block or serum anti-GM1 antibodies alone can be markers for patients with LMN syndromes that often improve after immunomodulating therapy.

Multifocal motor neuropathy (MMN) and motor syndromes with serum anti-GM1 antibodies.

A. Clinical Syndromes

The immune-mediated motor neuropathies are characterized by
asymmetric, slowly progressive weakness that most commonly begins in the arms. The age of onset is generally between 20 and 75. Men are affected somewhat more commonly than women. Motor findings include asymmetric weakness and variable degrees of atrophy. Patients with prominent conduction block may present with weakness in muscles with relatively normal bulk. Rarely, patients have had cranial nerve signs including external ophthalmoplegia and unilateral tongue weakness and atrophy. Some patients report paresthesias, but sensory signs are usually absent or clinically insignificant. In regions with normal strength tendon reflexes are often preserved. In areas of weakness, reflexes may initially be normal but can become reduced with progression of the disease. Fasciculations are not uncommon, and may add to diagnostic confusion between MMN and variants of amyotrophic lateral sclerosis (ALS) with only lower motor neuron signs. However, hyperreflexia and spasticity typical of ALS never occur in MMN.

Conduction block
Amount of block increases with
more proximal stimulation.

B. Electrodiagnostic Classification

Motor neuropathies can be initially subdivided on the basis of electrophysiologic data.

1) Patients with MMN have, by definition, focal block of nerve conduction along the course of motor, but not sensory, axons. Conduction block has been variably defined as a 15% to 50% reduction in the compound muscle action potential at proximal as compared with distal sites of stimulation. Conduction block may go undetected unless multiple segments in several nerves are tested. The finding of conduction block is most reliable when the change is focal and in a distal nerve segment other than regions of entrapment. Temporal dispersion and phase cancellation over long nerve segments should be ruled out. In MMN, motor conduction velocities and distal latencies are often unremarkable in regions without conduction block. Sensory studies are normal.

2) Patients with distal lower motor neuron (D-LMN) syndromes have, by definition, motor axon loss but no conduction block. However, they are clinically similar to MMN patients, with asymmetric, slowly progressive weakness that most commonly begins in the arms. In many patients with D-LMN syndromes electrophysiologic studies provide no definitive help in distinguishing immune-mediated motor neuropathy from untreatable motor neuron disease. However, temporal dispersion of compound muscle action potentials, F-wave abnormalities and other changes suggestive of demyelination occur in some patients with D-LMN syndromes. Evidence of mild segmental demyelination on nerve biopsy, while not diagnostic, may also provide a clue that a D-LMN syndrome is immune-mediated.

C. Anti-GM1 Antibodies

GM1 ganglioside
Using Co-GM1 methodology for measurement, high titer serum IgM anti-GM1 antibodies occur frequently in MMN (80% to 90% of cases) and in some patients with D-LMN syndromes. High titers of IgM anti-GM1 antibodies (above the normal range) are rare in most other disorders (< 1%) such as ALS and peripheral neuropathies. Anti-GM1 antibodies may be found in acute motor neuropathies without evidence of demyelination (35% to 40%). Very high titers of anti-GM1 antibodies (> 40,000) are often associated with IgM monoclonal antibodies in the serum. Optimal senstivity for MMN (85% to 90%) is obtained when serum is also tested for IgM binding to NP-9 (GM1 ganglioside in a membrane-like preparation of myelin lipids that also includes galactocerebroside, and cholesterol). Specificity of testing is evaluated by evaluating IgM binding to histone H3.

The occurrence of IgM binding to Co-GM1 or NP-9 in both MMN and D-LMN syndromes without conduction block suggests that these are pathogenically related disorders that are distinguishable from other immune neuropathies and ALS. As practical matters, when specific serum antibody binding to GM1 ganglioside (Co-GM1) and NP-9 is detected, additional electrodiagnostic studies searching for conduction block should be performed, and the patient considered for immunosuppressive therapy, if indicated by the degree of disability.

High titers (>1:1000) of selective serum IgG binding to GM1 ganglioside have strong specificity for chronic D-LMN syndromes and acute axonal motor neuropathies.

High titers of IgM binding to asialo-GM1 may also occur in MMN.

Overall, the finding of high-titer serum antibodies to Co-GM1 or NP-9:
1. provides strong independent support for the specific diagnosis of immune-mediated MMN and D-LMN syndromes
2. excludes other neurologic disorders such as polyneuropathies and ALS, and
3. indicates that immune-modulating treatment may be beneficial.

A major issue regarding clinical testing for anti-GM1 antibodies is differences in the methodology used in their measurement. Technical validation of methodology by identifying serums with high antibody titers is not sufficient. Laboratories must also document the sensitivity and specificity of anti-GM1 antibody testing methods by clinical correlation studies using serums from patients with motor syndromes.

In our laboratory the sensitivity of antibody (IgM vs Co-GM1 and NP-9) testing for MMN is now 85% to 90%. In general, high titers of serum IgM anti-GM1 antibodies should be detected in:
1. at least 80% of patients with multifocal motor neuropathy, and
2. less than 1% of patients with typical amyotrophic lateral sclerosis.

A common misleading practice is citation, in test reports, of statistics from the literature without clinical validation of the specific methods used in a laboratory. Laboratories that cannot provide correlation data, relating patient syndromes to results from their specific methodology, are not qualified to perform measurements of serum anti-GM1 antibodies as a clinical test.

D. Pathogenic Mechanisms

Although the exact role of anti-GM1 antibodies in producing disease has not been determined, there is growing evidence that they can be pathogenic. Immunization of rabbits with GM1 may result in the production of a neuropathy with similarities to MMN. Conduction block has been induced experimentally by the intraneural injection of serum immunoglobulin from patients with motor neuropathy and anti-GM1 antibodies. Anti-GM1 antibodies can alter K+ and Na+ currents in myelinated axons. Specific anatomical patterns of binding of some anti-GM1 antibodies to peripheral nerve, spinal cord, and motor neurons are consistent with clinical and electrophysiologic findings in MMN. Clinically, titers of anti-GM1 antibodies decline before improvement in strength after cyclophosphamide treatment. Finally, there is more GM1 ganglioside in myelin from motor nerves than from sensory nerves. This could render motor fibers more susceptible to attack by anti-GM1 autoantibodies and explain their selective involvement in MMN and D-LMN syndromes.

E. Treatment

Treatment with cyclophosphamide or human immune globulin (HIG) can produce useful functional improvement in patients with MMN.

Improvement in strength after treatment with HIG (for example 1 g/kg/day x 2 days) is common (50% to 70% of cases), but the length of benefit is variable, lasting from 2 weeks to 6 or more months. The dose and frequency of subsequent treatments is based on individual patient response.

1. The period of maximum improvement after HIG treatment should be monitored. Subsequent treatments should be given just before a relapse is expected. The minimum effective dose of HIG can be determined by sequentially reducing the subsequent HIG doses by 10% until a level is found that produces somewhat less benefit (length of time, or degree, of improved strength). The minimum dose that produced an optimal improvement is then used for long-term therapy.

2. Lack of improved strength after one, or at most two, treatments (total 2 to 3 g/kg each) should be considered a treatment failure. No further HIG should be used.

Although the side effects of HIG are usually benign, its great expense mandates objective documentation of any benefit, including quantitative muscle testing and functional assessment, to justify continued use. Further studies are required to document whether HIG treats the underlying pathogenic process in MMN or produces symptomatic benefit while allowing the underlying immune process to progress. Corticosteroid (Prednisone or Solumedrol) treatment is rarely helpful in MMN and may often exacerbate weakness.

Cyclophosphamide is the only immunosuppressive medication that has been reported to induce long-term benefit in many patients (50% to 80%) with MMN. Unfortunately, its toxicity, especially the increased risk of neoplasia with high cumulative life-time doses (>75 g), requires a careful analysis of the risk:benefit ratio in each patient. Therapeutic regimens should utilize doses of cyclophosphamide that are high enough to reduce anti-GM1 antibody titers by 60%, or more. We originally used an initial dose of 3 g/M2 over 8 days followed by chronic oral medication (100-150 - mg/day for 6 to 12 months). More recent experience suggests that 6 monthly treatments with intravenous cyclophosphamide (1g/M2), each preceded by two plasma exchanges, is equally effective, has fewer adverse effects and utilizes a 50% to 70% lower cumulative dose of drug. This regimen produces a sustained reduction in titers of serum anti-GM1 antibodies in approximately 60% to 80% of patients. Most patients in whom antibody titers are reduced show functional benefit. Remission usually persists for 1-3 years; after which, antibody titers often rise and weakness recurs. Retreatment may then be necessary.

Deciding whether to treat patients with D-LMN syndromes without electrodiagnostic evidence of demyelination may be difficult. High titers of serum IgM anti-GM1 antibodies are a useful indicator that a D-LMN syndrome may be immune- mediated and treatable. Evidence of demyelination on sural nerve biopsy may also be helpful in this regard. Measurable improvement in strength after treatment with HIG may provide support for further immunotherapy, with agents such as cyclophosphamide or further periodic HIG infusions.

2. Differential Diagnosis of Motor Neuropathies

The differential diagnosis of motor neuropathies includes motor neuron disorders, hereditary and acquired, on one hand, and demyelinating neuropathies, on the other.

Motor neuropathies and motor neuron disorders. Some motor neuropathies have been classified as ALS variants, with predominantly LMN signs and axonal changes on electrodiagnostic studies. Certain features can aid in the differentiation between motor neuropathies and ALS. Patients with motor neuropathies may have preserved reflexes in weak muscles, but overt spasticity and bulbar features are conspicuously lacking. This is in contrast to patients with ALS who often have prominent upper motor neuron and bulbar findings. The prolonged course that is often noted in patients with motor neuropathies also helps to differentiate their syndromes from typical ALS. Acquired motor neuropathies most often produce asymmetric weakness. This pattern is usually clinically distinct from the proximal symmetric weakness that characterizes most of the hereditary spinal muscular atrophies.

Several lower motor neuron syndromes have been described that are of uncertain etiology and could be disorders of the motor axon or cell body. A majority of patients with D-LMN syndromes have neither evidence of peripheral nerve demyelination nor serum anti-ganglioside antibodies. These patients tend to have more rapidly progressive weakness than is typical for the immune-mediated motor neuropathies. In contrast to typical ALS, many D-LMN patients never develop bulbar dysfunction. There are no reports of response to immunosuppressive treatment in D-LMN patients with neither demyelination nor serum autoantibodies. Some patients develop progressive asymmetric lower motor syndromes with predominant early weakness in proximal musculature (P-LMN syndromes). Characteristic features include late-age onset, male predominance (85%) and initial signs of weakness in the upper extremities (80%). Progression is slow. Weakness is often confined to one or two extremities for 3 to 5 years. Electrodiagnostic studies show only evidence of axonal loss. Some patients with P-LMN syndromes (30%) have selective serum antibody binding to GA1 ganglioside. However, there is no evidence that P-LMN syndromes respond to immunosuppressive treatment.

Monomelic amyotrophy, a syndrome that affects mainly young (15 to 25 years) males (80%), presents with weakness of the distal musculature of one upper extremity that progresses for 1 to 2 years and then remains stable. Occasional patients develop weakness in the opposite limb, mild sensory symptoms or tremor. Electrodiagnostic studies show denervation in the affected limb. There are no associated serum antibodies.

Rare patients with paraneoplastic LMN syndromes have been reported. The best described of these is a subacute motor neuronopathy associated with lymphomas, such as Hodgkin’s disease. Progressive, asymmetric weakness develops, most severely in the legs, at times when the neoplasm is in remission or during irradiation. The weakness is rarely severe, and often stabilizes or improves over a period of months to years. Pathological studies show a loss of motor neurons in the ventral horn of the spinal cord and some involvement of sensory tracts. LMN involvement has also been described as an occasional part of the paraneoplastic encephalomyelitis and sensory ganglionopathy syndromes that occur in association with anti-Hu antibodies. There is no clear evidence that there is an increased incidence of paraneoplastic "typical" ALS syndromes, with upper and lower motor neuron involvement. A few patients with ALS-like syndromes and neoplasms, including renal cell, lung and lymphoma, have been reported to improve or stabilize after treatment of the cancer.

Immune demyelinating Neuropathies. Although MMN and CIDP are both demyelinating neuropathies, the differences in their clinical, electrophysiological and immunologic features are more prominent than their similarities. MMN commonly presents with distal asymmetric weakness while in CIDP, proximal symmetric weakness is a more common finding. The remitting and relapsing course that may occur in CIDP is uncommon in the motor neuropathies. Patients with MMN rarely have significant sensory symptoms while in CIDP, sensory signs are the rule. Electrophysiological testing may show conduction block in both conditions. However, other features of demyelination such as prolonged distal latencies and slowed conduction velocities are more prominent in CIDP. Abnormalities in sensory nerve conduction studies are usually seen in CIDP, but not in MMN, unless complicated by another disease process. The spinal fluid examination shows markedly increased protein concentration in the majority of cases of CIDP while this change is rare in patients with MMN. High titer anti-GM1 antibodies as well as more specific patterns of autoantibody reactivity (see above) are common in MMN. In CIDP anti-GM1 antibodies are unusual. Serum autoantibody binding to tubulin is more common. Finally, differences in the frequency of therapeutic response to prednisone and plasma exchange (common in CIDP, but rare in motor neuropathies) define a practical difference in the management of the two disorders.

Other immune-mediated demyelinating neuropathies have more sensory involvement and are rarely confused with MMN. Anti-MAG neuropathies are often associated with weakness, but sensory loss is most often the presenting, and disabling, feature of the disease. Neuropathies with anti-sulfatide antibodies and GALOP syndrome have even more predominant sensory involvement. POEMS syndrome may produce severe weakness but this is accompanied by prominent sensory loss and systemic signs.

3. Illustrative cases

CASE 1: A 42 year old woman with progressive weakness in the right hand for one year was referred for possible motor neuron disease. During the next year the hand weakness became more severe and also appeared in new areas of the body, first the left hand and then the left foot. Physical examination revealed asymmetric weakness that was predominantly distal. The thenar eminence in the left hand was severely weak but showed no wasting. Tendon reflexes were 1+ throughout. Sensory testing showed only a minimal reduction in vibratory sensation at the toes. Electrodiagnostic studies showed motor conduction block in the median nerves bilaterally between the elbow and wrist. Nerve conduction velocities were normal. Sensory nerves were normal. Serum IgM anti-GM1 antibodies were present in high titer (3,900; normal <600).

During the next 10 months there was progressive weakness and loss of reflexes despite high dose prednisone therapy and 11 treatments with plasma exchange. Six monthly treatments with two plasma exchanges followed by intravenous cyclophosphamide (1 gm/M2 ) were followed by improvement, beginning after 3 to 4 months, and progressing to nearly normal strength over the next year. She remained stable, off all medications, for 3 years. She then noted mild recurrent weakness in the right hand in a distribution similar to that at disease onset 5 years before.

COMMENT: Asymmetric weakness, developing distally in an arm or hand, is the most common pattern of early involvement in MMN. Reflexes are often normal early in the disease course. Significant sensory signs are rare, but patients occasionally note symptoms such as paraesthesias, or even abnormal taste sensations. Prednisone treatment is rarely effective, and is often associated with rapid exacerbations of weakness. The decision to use cyclophosphamide was only made when 1. it became clear that the patient had developed significant disability, and 2. there were clear signs that an immune disorder was present, including conduction block and high serum titers of IgM anti-GM1 antibodies. Improvement in strength after cyclophosphamide begins late, often 3 to 6 months after beginning therapy, and continues for up to a year after the end of the treatment course.

CASE 2: A 52 year old male noted a right foot drop. During the next 6 months weakness and cramps became progressively worse in the right leg and also developed in the left leg. On examination there was asymmetric weakness, predominantly in the legs. Muscle tone was normal. Cranial nerves were normal. Tendon reflexes were absent at the right ankle, but 2+ elsewhere. Sensation was normal. Electrodiagnostic testing showed denervation in thoracic paraspinous muscles and both lower extremities. Nerve conduction velocities were normal. No serum anti-GM1 antibodies were detected.

COMMENT: This patient has relatively rapidly progressive weakness with evidence of involvement of lower, but not upper, motor neurons. The clinical course is more rapid than that usually seen in MMN. The most appropriate diagnostic categories would be the older term, “progressive muscular atrophy”, or, descriptively, “lower motor neuron syndrome”. Such patients may not develop bulbar, or upper motor neuron, signs and never meet diagnostic criteria for ALS. Denervation in the thoracic paraspinous muscles is more suggestive of a motor neuron disease than a motor neuropathy. With no evidence of demyelination or serum antibodies to suggest an immune etiology for the syndrome, it is likely that this patient will have continued progression of weakness that does not respond to immunosuppressive treatments.

CASE 3: A 55 year old man was referred for possible immunosuppression to treat a lower motor neuron syndrome with anti-GM1 antibodies. He had noted increasing difficulty climbing stairs and arising from a chair for 10 to 15 years. In recent years his speech had become slurred and he needed more time to eat. On general examination there was mild gynecomastia. Neurological testing showed weakness and fasciculations of the tongue and face. The tongue showed severe atrophy. Moderate symmetric, proximal weakness was present. Tendon reflexes were difficult to elicit. Sensation was reduced to all modalities distally in the feet. Electrodiagnostic testing showed chronic denervation, most prominent in the face, tongue, and proximal muscles. Repeat anti-GM1 antibody testing showed a pattern of polyreactive serum IgM binding to GM1 ganglioside and to histone H3 at titers of about 1,500.

COMMENT: The patients weakness was proximal and symmetric, more typical of inherited motor neuron disorders than of acquired motor neuropathies. Although the patient had high titers of anti-GM1 antibodies, the pattern of binding was polyreactive. Polyreactive antibodies are not specific for immune motor syndromes, and are also found in 3% to 5% patients with ALS and adult-onset spinal muscular atrophies. Further testing revealed an excessive number of trinucleotide repeats in the androgen receptor, a finding consistent with X-linked hereditary bulbo-spinal muscular atrophy. This case emphasizes that determination of the specificity of anti-GM1 antibodies helps to determine their clinical relevance. The polyreactive antibodies in a patient with features atypical of a motor neuropathy were not, of themselves, an indication for immunosuppressive therapy.


Kornberg AJ, Pestronk A. Chronic motor neuropathies: Diagnosis, therapy, and pathogenesis. Ann Neurol 1995; 37(S1):S43-S50.
Descriptions of lower motor neuron syndromes and motor neuropathies.

Lopate G, Pestronk A. Chronic immune demyelinating neuropathies. Seminars in Neurology 1994;14:131-136.
Differential diagnosis of demyelinating neuropathies, including MMN

Parry G. Motor neuropathy with multifocal conduction block. In: Dyck PJ, Thomas PK, Griffin JW, Low PA, Poduslo JF, eds. Peripheral Neuropathy. 3rd ed. Philadelphia: W.B. Saunders Company 1993 pp 1518-1524.
MMN with an emphasis on the electrodiagnostic features

Pestronk, A., Choksi, R. Multifocal motor neuropathy: Serum IgM anti-GM1 ganglioside antibodies in most patients detected using covalent linkage of GM1 to ELISA plates. Neurology 1997;49:1289-1292.

Pestronk, A, Choksi R, Blume, G, Lopate, G. Multifocal motor neuropathy: Serum IgM binding to a GM1-ganglioside containing lipid mixture but not to GM1 alone. Neurology 1997;48:1104-1106.

Pestronk A, Lopate G, Kornberg AJ, et al. Distal lower motor neuron syndrome with high-titer serum IgM anti-GM1 antibodies -improvement following immunotherapy with monthly plasma exchange and intravenous cyclophosphamide. Neurology 1994;44:2027-2031.
Treatment for lower motor neuron syndromes and MMN using lower cumulative doses of intravenous cyclophosphamide.

Pestronk A, Chaudhry V, Feldman EL, et al. Lower motor neuron syndromes defined by patterns of weakness, nerve conduction abnormalities, and high titers of antiglycolipid antibodies. Ann Neurol 1990;27:316-326.
Detailed description of different lower motor neuron syndromes and associated autoantibodies.

Takigawa T, Yasuda H, Kikkawa R, Shigeta Y, Saida T, Kitasato H. Antibodies against GM1 ganglioside affect K+ and Na+ currents in isolated rat myelinated nerve fibers. Ann Neurol 1995;37:436-442.
Pathogenic effects of anti-GM1 antibodies.

Return to
Return to Motor Syndromes
Return to Neuromuscular Home Page