CIDP = chronic inflammatory demyelinating polyradiculoneuropathy, HNPP = hereditary neuropathy with liability to pressure palsy, CMT = Chercot-Marie-Tooth disease, HIV = human immunodeficiency virus.

* Include diabetes mellitus, uremia, thyroid disorders.

Chronic Demyelinating Polyneuropathies

In most peripheral polyneuropathies, it is often possible to define the predominant pathophysiologic mechanism, based on electrophysiologic and pathologic features, as being either primarily axonal or demyelinating. In demyelinating polyneuropathy, it is also useful to distinguish between neuropathies with segmental (multifocal) versus uniform slowing, based on electrophysiologic studies (see electrodiagnosis). Multifocal or segmental demyelinating polyneuropathies are almost always acquired, while uniform demyelinating polyneuropathies are typically hereditary.

The causes of chronic axonal neuropathies are abundant, while chronic demyelinating polyneuropathies have a fairly restrictive differential diagnosis (Table C18-2). Many acquired demyelinating polyneuropathies are immune in nature and respond to immunosupression or immunomodulation, while most axonal polyneuropathies are metabolic or toxic in nature. Since the differential diagnosis of chronic acquired demyelinating polyneuropathies is quite limited, the diagnostic work-up for patients with such entities is much less laborious and is quite different from that of patients with axonal neuropathies (Table C18-3).

Table C18-2 Common Causes of Chronic Demyelinating Peripheral Polyneuropathy

Hereditary (uniform slowing)

1. HMSN I and III (CMT1 and 3)

2. HNPP ^*

3. Congenital hypomyelinating neuropathy

5. Cerebrotendinous xanthomatosis

6. Refsum disease

7. Tangiers disease

CIDP = chronic inflammatory demyelinating polyradiculoneuropathy; HIV = human immunodeficiency virus; GM₁ = ganglioside M₁; MAG = myelin-associated glycoprotein; MGUS = monoclonal gammopathy of unknown significance; Ig = immunoglobulin; POEMS syndrome = polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, and skin changes; HMSN = hereditary motor and sensory neuropathy; CMT = Charcot-Marie-Tooth disease; HNPP = hereditary neuropathy with liability to pressure palsy.

* May have multifocal slowing also, usually across common entrapment sites.

Table C18-3 Recommended Work-Up of Chronic Acquired Demyelinating Peripheral Polyneuropathy

BUN = blood urea nitrogen; CBC = complete blood count; CSF = cere-brospinal fluid; GM1 = ganglioside M1; HIV = human immunodeficiency virus; MAG = myelin-associated glycoprotein.

* Serum immunofixation is often necessary because routine serum protein electrophoresis may miss patients with a small amount of circulating paraprotein (M-protein).

Chronic Inflammatory Demyelinating Polyradiculoneuropathy

Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is the prototype of all chronic acquired demyelinating polyneuropathies. It is an autoimmune disorder of the peripheral nervous system that affects individuals at any age and may be relapsing and remitting or slowly progressive usually over several months. Proximal and distal symmetrical weakness is the most common manifestation. Many patients have also numbness and paresthesias, usually of the feet and hands. Generalized areflexia is very common while some patients have only hyporeflexia or distal areflexia.

Patients with CIDP must be distinguished from patients with the more prevalent acquired axonal peripheral polyneuropathies. CIDP should also be separated from hereditary polyneuropathies, particularly those with demyelinating features such as CMT1, CMTX, and CMT3. CIDP should also be distinguished from other polyradiculopathies, such as meningeal carcinomatosus, Lyme disease, or sarcoidosis. When predominantly motor, CIDP may mimic neuromuscular junction disorders (such as the Lambert-Eaton myasthenic syndrome), motor neuron disorders, and myopathies.

Chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a diagnosis of pattern recognition, based on clinical manifestations, EDX, cerebrospinal fluid examination, laboratory tests appropriate to the specific clinical situation, and, occasionally, results from nerve biopsy. The American Academy of Neurology defined criteria for the diagnosis of CIDP (Table C18-4). Four features are set as the basis of diagnosis: clinical, electrodiagnostic, pathologic, and cerebrospinal fluid studies. These are further divided into mandatory, supportive, and, where appropriate, exclusion. Mandatory features are those required for diagnosis and should be present in all definite cases. Supportive features are helpful in clinical diagnosis but by themselves do not make a diagnosis. Exclusion features strongly suggest alternative diagnoses.

Table C18-4 American Academy of Neurology Criteria for Diagnosis of Chronic Inflammatory Demyelinating Polyneuropathy

II. Electrodiagnostic Studies

A. Mandatory

Nerve conduction studies including studies of proximal nerve segments in which the predominant process is demyelination

Must have three of four:

1. Reduction in conduction velocity (CV) in two or more motor nerves:

a. <80% of lower limit of normal (LLN) if amplitude >80% of LLN

b. <70% of LLN if amplitude <80% of LLN

2. Partial conduction block ^* or abnormal temporal dispersion^† in one or more motor nerves: either peroneal nerve between ankle and below fibular head, median nerve between wrist and elbow or ulnar nerve between wrist and below elbow

3. Prolonged distal latencies in two or more nerves:

a. >125% of upper limit of normal (ULN) if amplitude >80% of LLN

b. >150% of ULN if amplitude <80% of LLN

4. Absent F waves or prolonged minimum F-wave latencies (10 to 15 trials) in two or more motor nerves:

a. >120% of ULN if amplitude >80% of LLN

b. >150% of ULN if amplitude <80% of LLN

B. Supportive

1. Reduction in sensory CV <80% of LLN

2. Absent H reflexes

III. CSF Studies

A. Mandatory

1. Cell count <10/mm³ if HIV-seronegative or <50/mm³ if HIV-seropositive

2. Negative VDRL

B. Supportive

Elevated protein

Diagnostic categories. Definite: Clinical A and C, Electrodiagnostic A, CSF A, and Pathology A and C. Probable: Clinical A and C, Electrodiagnostic A, and CSF A. Possible: Clinical A and C and Electrodiagnostic A.

* Criteria suggestive of partial conduction block: >20% drop in area or amplitude with <15% change in duration between proximal and distal sites.

† Criteria for abnormal temporal dispersion and possible conduction block: >20% drop in area or amplitude between proximal and distal sites with >15% change in duration between proximal and distal sites and. These criteria are only suggestive of partial conduction block as they are derived from studies of normal individuals. Additional studies, such as stimulation across short segments or recording of individual motor unit potentials, are required for confirmation.

‡ Demyelination by either electron microscopy (>5 fibers) or teased fiber studies (>12% of 50 teased fibers, minimum of four internodes each, demonstrating demyelination/remyelination).

Peripheral Polyneuropathy and Monoclonal Gammopathy of Undetermined Significance

The prevalence of monoclonal gammopathy of undetermined significance (MGUS) increases with age. It is present in 1% of patients older than 50 years of age and in 3% of patients older than 70 years of age. This entity must be distinguished from the more malignant myeloproliferative disorders, such as multiple myeloma, by obtaining complete blood count (CBC), calcium, blood urea nitrogen (BUN)/creatinine, a skeletal survey, and, at times, a bone marrow aspirate. Table C18-5 lists both the criteria needed to confirm the diagnosis of MGUS and its common characteristics. Although MGUS is relatively benign and is commonly asymptomatic, follow-up reveals that malignant myeloproliferative disorders will develop in up to one-third of these patients within 20 years. A good indication for this malignant transformation is a rising M-protein (paraprotein) value, especially one greater than 3 g/dL. Thus, a regular follow-up of the paraprotein value is warranted in all patients with MGUS.

Table C18-5 Monoclonal Gammopathy of Unknown Significance (MGUS)

Criteria for Diagnosis	Characteristics
1. Paraprotein (M-protein) value less than 3 g/dL 2. No lytic bony lesions 3. Plasma cell in bone marrow less than 5% 4. No anemia, hypercalcemia, or renal failure

Common monoclonal type: IgM or IgG

Common light chain: kappa

Urine monoclonal protein: rare

Up to 10% of patients with peripheral polyneuropathy have a monoclonal protein; this is significantly higher than the prevalence in the general population (1–3%). The paraprotein is commonly of the IgG or IgM class and less often of the IgA class. Almost half of patients with IgM-associated neuropathy have elevated serum antibody titers to myelin-associated glycoprotein (anti-MAG). However, patients with elevated anti-MAG antibody do not all have a detectable IgM paraprotein on immunofixation.

The polyneuropathies associated with paraproteinemia (e.g., MGUS, myeloma) are heterogeneous; they can be axonal or demyelinating, sensory or sensorimotor. Their characteristics correlate poorly with the class of abnormal paraprotein. Certain important points need to be emphasized:

1. When it is demyelinating in nature, MGUS-associated polyneuropathy is indistinguishable from CIDP, both clinically and electrophysiologically, except by the presence of the serum paraprotein.

2. Immunoglobulin M-associated neuropathy is predominantly sensory and ataxic, and often is demyelinating.

3. Anti-MAG-associated polyneuropathy, similar to IgM-associated neuropathy, is frequently predominantly sensory, ataxic, and demyelinating.

4. A subgroup of patients with myeloma, those with osteosclerotic myeloma, commonly present with a chronic demyelinating polyneuropathy. Typically, there is a single or multiple sclerotic bone lesion that harbors monoclonal plasma cells. This disorder also is termed POEMS (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, and skin changes) because of other frequently associated features (hepatosplenomegaly, gynecomastia, atrophic testicles, hyperpigmentation, hypertrichosis, and clubbing).

There is continuous debate regarding the various acquired demyelinating polyneuropathies, namely CIDP, MGUS neuropathy, anti-MAG-associated neuropathy, and multifocal motor neuropathy with conduction block. Our current knowledge of the exact etiology and pathogenesis of these immune disorders is lacking. Figure C18-2 reveals a schematic representation of the significant overlap between all these disorders. Apart from the presence of a monoclonal protein, the clinical and EDX features of CIDP – with or without MGUS – are quite similar. However, as is shown in Table C18-6, there are certain features in the presentation and clinical course that tend to help differentiate between these disorders.

Figure C18-2 Spectrum of acquired demyelinating polyneuropathies.

Table C18-6 Distinctive Features for Differentiating Between Chronic Inflammatory Demyelinating Polyradiculoneuropathy (CIDP) Without and With MGUS

Feature	CIDP Without MGUS	CIDP With MGUS
Age	Relatively younger	Relatively older
Course	Progressive or relapsing	Frequently progressive
Neuropathy	Predominantly motor	Predominantly sensory with ataxia
Clinical deterioration	More rapid	Slow indolent
Functional impairment	Moderate to severe	Mild
Spontaneous improvement	Common	Rare
Response to therapy	Good	Less responsive

The treatment of patients with MGUS-associated polyneuropathy depends on its clinical presentation. Patients with sensory symptoms only, particularly the elderly, may be treated symptomatically with drugs that alter neuropathic pain. Plasma exchange is effective, particularly in neuropathies with IgG and IgA type MGUS, but should be reserved for patients with significant motor weakness or ataxia. Intravenous gamma globulin is useful particularly in patients with a CIDP-MGUS presentation. Prednisone, rituximab, azathioprine, chlorambucil, and cyclophosphamide have resulted in benefit.

Electrodiagnosis

In a patient with suspected peripheral polyneuropathy, the EDX study:

1. Provides an unequivocal diagnosis of peripheral polyneuropathy.

2. Defines the anatomic distribution of a neuropathy, as a single mononeuropathy, multiple mononeuropathies, or a generalized peripheral polyneuropathy.

3. Excludes mimickers of polyneuropathies such as bilateral L5 and S1/S2 radiculopathies, bilateral tarsal tunnel syndromes and, occasionally, bilateral carpal tunnel syndromes, or distal myopathies.

4. Establishes the type of fiber(s) affected (sensory, motor, or both).

5. Estimates the chronicity and activity of the process.

6. Most importantly, identifies its primary pathologic process (demyelination or axonal loss).

Thus, at the completion of an EDX test, the clinician should be able to better characterize the polyneuropathy and classify its pathophysiology. This helps establish a relatively short differential diagnosis and work-up aimed at identifying the cause of the neuropathy and planning its management (see Table C18-1).

Analyzing conduction times (velocities and latencies), as well as CMAP amplitude, area and duration, is an essential exercise in the EMG laboratory for establishing the primary pathologic process of a polyneuropathy. In most situations, the polyneuropathy falls in one of the following categories based on one of two primary nerve dysfunctions: the axon or its supporting myelin. Occasionally, such as in very mild polyneuropathies or in severe situations associated with absent responses, it may be difficult to establish the primary pathology based on EDX studies.

• Primary axonal polyneuropathies (axonopathies) affect the axon primarily and produce a length-dependent dying-back degeneration of axons. The major change on NCS is a decrease of the CMAP and SNAP amplitudes, more marked in the lower extremities. In contrast, conduction times (velocities, distal latencies, and F wave minimal latencies) are normal. Sometimes, there is a slight slowing of distal latencies, conduction velocities and F wave minimal latencies when the polyneuropathy is advanced (Figure C18-3). This is explained by the fact that the loss of axons is distributed in a random fashion, which results in survival of some fast-conducting fibers (Figure C18-4B). Figure C18-5 reveals the theoretical distribution of conduction velocity in motor nerves of healthy patients and patients with axonal neuropathy. Unless there is selective loss of largely myelinated, fast-conducting fibers, the axonal loss is indiscriminate, resulting in survival of some fast-conducting fibers and leading to normal velocities. It is only when axonal loss is severe, surpassing 75 to 80% of the total population of axons, that slight slowing of velocities occurs. In these situations, conduction velocities should be no less than 70% of the lower limit of normal.

• Primary demyelinating polyneuropathies (myelinopathies) are, however, characterized by significant slowing of conduction times (velocities, distal latencies, and F wave latencies) because the pathologic process results in myelin disruption (segmental and paranodal demyelination) which impedes saltatory conduction. Commonly, the CMAP amplitudes are relatively preserved distally, although conduction blocks and dispersion are common, mainly in the acquired forms (such as CIDP). With distal stimulation, the CMAP is mildly reduced in amplitude because of temporal dispersion and phase cancellation. The distal latency is slowed (>130% of upper limit of normal) because of demyelination. With more proximal stimulation, the CMAP is much lower in amplitude, which results from temporal dispersion and conduction block along some fibers. The proximal conduction velocities are markedly slow (<70% of lower limit of normal) because of increased probability for the action potentials to pass through demyelinated nerve segments (Figure C18-4C). If untreated, CIDP is often progressive and debiltating with secondary axonal loss (Figure C18-6).

Figure C18-3 Peroneal motor conduction nerve conduction studies in a control (A) and in an age-matched patient with axonal polyneuropathy due to chronic alcoholism (B). Note the significant decrease in CMAP amplitudes in (B) compared to (A), while there is only slight slowing of distal latencies and conduction velocities.

Figure C18-4 Computerized model of peripheral motor nerve in normal nerve (A), axonal degeneration (B), and segmental demyelination (C).

(Adapted from Albers JW. Inflammatory demyelinating polyradiculoneuropathy. In: Brown WF, Bolton CF, eds. Clinical electromyography. Boston, MA: Butterworth-Heinemann, 1989.)

Figure C18-5 Computer simulation of the effect on the distribution of conduction velocities of a loss of 75% of the motor units. (A) Normal. (B) Abnormal.

(From Osselton JW et al., eds. Clinical neurophysiology, EMG, nerve conduction and evoked potentials. Oxford: Butterworth-Heinemann, 1995.)

Figure C18-6 A 75-year-old man developed an indolent distal sensory loss and mild ataxia beginning in the summer of 2003. He was diagnosed with CIDP in November 2003 but was not treated. He continued to worsen steadily and, in December 2004, became wheelchair bound with complete areflexia, severe sensory loss, and severe proximal weakness. His EDX studies showed significant worsening. The ulnar motor conduction study in November 2003 (A) compared to December 2004 (B). Note the worsened latencies and amplitudes and evidence of conduction block in December 2004.

Multiple criteria have been set for the diagnosis of CIDP and are aimed at distinguishing the primary demyelinating polyneuropathy from the primary axonal polyneuropathy. Criteria proposed by Cornblath and Asbury were adopted by consensus to apply to patients with suspected CIDP (see Report from an Ad Hoc subcommittee, 1991). Table C18-7 reveals common nerve conduction criteria used to identify the acquired demyelinating polyneuropathies.

Table C18-7 Electrophysiologic Criteria for Acquired Demyelinating Polyneuropathy

Criteria	Albers and Kelly	Asbury and Cornblath
*Required* 1. Conduction block/w temporal dispersion

≥3 Criteria

≥3 Criteria

<90% LLN if CMAP >50% LLN

<80% LLN if CMAP <50% LLN

in ≥ 2 nerves

<80% LLN if CMAP >80% LLN

<70% LLN if CMAP <80% LLN

3. Slowing of MDL

in ≥2 nerves

>115% ULN if CMAP >LLN

>125% ULN if CMAP <LLN

in ≥ 2 nerves

>125% ULN if CMAP >80% LLN

> 150% ULN if CMAP <80% LLN

>120% ULN if CMAP >80% LLN

>150% ULN if CMAP <80% LLN

MCV = motor conduction velocity; MDL = motor distal latency; ULN = upper limit of normal; LLN = lower limit of normal; CMAP = compound muscle action potential.

* Conduction block = >30% drop in CMAP amplitude between distal and proximal stimulations.

† Conduction block = >20% drop in CMAP area or amplitude between distal and proximal stimulations with <15% increase in CMAP duration; temporal dispersion = >20% drop in CMAP area or amplitude between distal and proximal stimulations with >15% increase in CMAP duration.

Adapted from Brown WF. Acute and chronic inflammatory demyelinating neuropathies. In: Brown WF, Bolton CF, eds. Clinical electromyography. Boston, MA: Butterworth-Heinemann, 1993, pp. 533–559; Albers JW, Kelly JJ. Acquired inflammatory demyelinating polyneuropathies: clinical and electrophysiologic features. Muscle Nerve 1989;12:435–451; Cornblath DR. Electrodiagnostic abnormalities in Guillain-Barré syndrome. Ann Neurol 1990;27(suppl):S17–S20.

Conduction block is defined as the loss of CMAP amplitude and area with proximal stimulation. The diagnosis of conduction block on nerve conduction studies requires a special detailed analysis of several CMAP parameters, including amplitude, duration, and area. In general, physiologic temporal dispersion, due to interphase cancellation, is length-dependent (i.e., more prominent in longer than shorter nerves, and in tall versus short subjects). This results in some loss in amplitude and area between distal and proximal stimulations in normal subjects, and even greater loss in long nerves and tall subjects.

Based on this, when the diagnosis of conduction block is being considered and criteria are being established in EMG laboratories, special thought should be given to the following:

• Criteria for conduction block should not be universal but should be individualized to the specific nerve and the specific segment of the nerve being studied.

• During analysis, special attention should be paid to the duration of the CMAP while a drop in amplitude is being evaluated. The diagnosis of conduction block can be difficult to confirm if significant (pathologic) temporal dispersion is present. This usually results in significant (>15%) prolongation of CMAP negative peak duration. Measurement of the negative peak area, which requires the use of computerized equipment, is extremely useful in these situations. Many criteria have been advocated for the diagnosis of conduction block, some based on normal controls and others on patients with demyelinating neuropathy. Table C18-8 reveals a practical approach for the diagnosis of conduction block and Figure C18-7 shows a practical algorithm in the diagnosis of conduction block, temporal dispersion or both.

Table C18-8 Electrodiagnostic Criteria of Conduction Block