The nerve conduction studies provide the information about the involvement of axon, myelin, or both. The characteristic findings in purely axonal lesions are similar to those in axonotmesis. The structural changes in interrupted nerve fibers can be classified into early axonal changes, clearance of myelin, the Schwann cell responses, and preparation for regeneration and late phase of Schwann cell atrophy and fibrosis (Bunge and Griffin, 1992) (Fig. 5.20). The nerve conduction velocity remains nearly normal throughout the Wallerian degeneration (Wilbourn, 1977). Distal latency and conduction velocities are preserved as long as a response can be obtained. In Wallerian degeneration of peripheral nerve, a length-dependent reduction in action potential takes place occurring earlier in shorter than longer nerves. The amplitude of compound muscle action potential (CMAP) declines earlier than sensory nerve action potential (SNAP). A near-normal conduction velocity is attributed to the loss of myelin sheath being secondary to axonal damage. Immediately after section of axon, the CMAP amplitude on motor nerve stimulation, SNAP on sensory nerve stimulation and distal latency remain normal up to 1 week. Needle EMG is normal although the recruitment pattern of motor unit potentials (MUPs) is reduced depending upon the degree of weakness. The CMAP and SNAP amplitudes progressively decrease and become unrecordable later. EMG changes of denervation appear by 1–3 weeks depending upon the proximity of the muscles to the site of nerve injury. On electromyography, the initial changes are prolonged insertional activity, positive sharp waves, fibrillations, and complex repetitive discharges. With partial axonal damage, the recruitment pattern is reduced, whereas no MUPs can be recruited in complete lesions. Initially the configuration of MUPs is normal; whereas in the succeeding months, the MUPs become polyphasic, high amplitude with long duration, and reduced recruitment (Table 5.18).

Conduction block and temporal dispersion are important neurophysiological findings in segmental demyelinating neuropathies. Conduction block is defined as the drop in the amplitude or area of CMAP on proximal stimulation compared to distal (Fig. 5.21A). A drop of 20–50% of the negative peak has been suggested for the diagnosis of conduction block (Brown and Feasby, 1984; Oh et al., 1994). Even in normal subjects some drop of CMAP amplitude on proximal stimulation occurs. In certain nerves such as median, ulnar, and peroneal, normally there may be a drop in CMAP amplitude up to 30% on proximal stimulation; whereas in posterior tibial nerve it is up to 41% (Oh et al., 1994). The stimulation of posterior tibial nerve in the popliteal fossa is technically difficult and may account for the greater drop of CMAP amplitude. In acquired demyelinating neuropathies, there is variable degree of demyelination, which is patchy and focal. This results in dispersion of CMAP due to phase cancellation. The negative and positive phases of action potential overlap due to variable conduction velocity in demyelinated fibers, leading to temporal dispersion. Such an amplitude drop may be erroneously diagnosed as conduction block. For interpreting the conduction block, therefore, the duration of CMAP should also be considered. A reduction of CMAP amplitude by 20% if its duration is normal; and 30% if CMAP duration is prolonged has been suggested (Brown and Feasby, 1984). To obviate this difficulty, the area measurement of CMAP employing a computer program has been recommended for diagnosis of conduction block. Peakto-peak measurement of CMAP amplitude has also been used for the diagnosis of conduction block, but the negative peak measurements are more appropriate. On proximal stimulation, the duration of CMAP increases slightly compared to that on distal stimulation, but an increase exceeding 20% is indicative of dispersion (Fig. 5.21B, Albers et al., 1985). The degree of reduction of CMAP amplitude if 20% was considered possible conduction block (Cornblath, 1990) and more than 30% as probable conduction block (Gilliatt, 1966). On histopathological correlation, the patients with even 20% reduction in CMAP amplitude were found to have definite evidence of demyelination. Hence a 20% reduction of the amplitude of proximal CMAP has been regarded as an indicator of segmental demyelination. For measurement of dispersion, although the total duration of CMAP may be measured but the duration of the negative peak is commonly employed (Oh et al., 1994).

Although the nerve conduction and EMG studies are useful in diagnosis of peripheral neuropathies, classifying them and even detecting the subclinical involvement, these studies however have certain limitations. It is important to appreciate these limitations not to undermine their clinical value, but to know how much to expect and where additional investigations are necessary. The limitations of nerve conduction studies are summarized in Table 5.23.

HMSN-III is defined as a severe demyelinating or hypomyelinating type of neuropathy with extremely slow motor nerve conduction velocities and delayed motor development. In HMSN-III the conduction velocity is much slower (<10m/s) compared to HMSN-I (Fig. 5.22). In a study comprising of 11 patients with HMSN-III, the median and ulnar nerve conduction velocity was <6m/s in all except one patient. Marked temporal dispersion without conduction block was present in all the patients. Uniform slowing in adjacent motor nerves was consistent with other studies of hereditary neuropathies although pronounced temporal dispersion may make HMSN-III more difficult to distinguish from acquired neuropathies (Benstead et al., 1990). Nerve conduction studies may be used as a screening test for detecting the subclinical cases in the families of HMSN-III. Originally, the disorder was thought to be autosomal recessive, the majority of the cases are sporadic or due to dominant mutations. A comparison of HMSN-type I and III is summarized in Table 5.26.

It is an autosomal recessive neuropathy manifesting in early childhood with progressive weakness. By adolescence these patients become bedridden. Nerve conduction is slowed (20–30m/s) and CSF protein normal. Nerve histology reveals loss of myelinated fiber, hypomyelination, and onion bulb formation. Numerous genetic loci have been detected and are mentioned in Table 5.24.

It is an autosomal dominant disease of peripheral nerves leading to increased susceptibility to mechanical traction or compression. Patients with HNPP manifest with recurrent episodes of isolated mononeuropathies. In order of frequency it results in common peroneal nerve palsy (foot drop) ulnar, radial, and median neuropathy; and brachial plexopathy. Attacks are usually precipitated sometimes even by slight compression or minor trauma. The weakness is sudden, painless, and recovers in days to weeks completely. The uncommon manifestations are transient positional sensory symptoms, chronic sensory polyneuropathy, and progressive mononeuropathy, which may be associated with pes cavus. Nerve conduction studies reveal prolonged distal motor latency, mild slowing of conduction velocity of forearm segment of median and ulnar nerve, focal slowing of ulnar and peroneal nerve at the usual sites of compression and diffuse reduction of SNAP amplitude (Dubourg et al., 2000, Mouton et al., 1999). Nerve conduction abnormalities (reduction of SNAP amplitude and prolonged terminal latency) have been described in asymptomatic carriers (Infante et al., 2001).

CMT-II constitutes one-third of all autosomal dominant CMT disease. It is a heterogenous group of axonal neuropathies with normal or slightly slowed nerve conduction velocity. Clinically, it resembles CMT-I but has a later age of onset (second decade or later), lower frequency of foot and spinal deformity, tremor, upper limb involvement, and generalized areflexia. The nerves are not thickened. About 20% patients are asymptomatic. The subtypes of CMT-II include CMT-IIA (chromosome 1p35–36), CMT-IIB (chromosome 3q13–22), CMT-IIC (Chromosome 12q24), and CMT-IID (chromosome 7p13; Saito et al., 1997). CMT-IIB has prominent sensory loss with ulcerations (DeJonghe et al., 1997). CMT-IIC is the most severe form and manifests with vocal cord, intercostal and diaphragmatic muscle weakness along with distal neuropathy resulting in respiratory failure and shortened lifespan (Dyck et al., 1994; Klain et al., 2002). Motor nerve conduction velocity may be normal or mildly slowed; SNAPs are either absent or reduced in amplitude.

Hereditary sensory autonomic neuropathy (HSAN) is a rare disorder, which manifests with prominent sensory loss and autonomic dysfunction without significant weakness. Sensory loss in HSAN predisposes to unnoticed recurrent trauma resulting in Charcot joints, nonhealing ulcers, infections, osteomyelitis, and acral mutilation giving the term acrodystrophic neuropathy (Fig. 5.27). Based on clinical features and type of sensory neuron involvement the HSAN is classified into five groups which are summarized in Table 5.27. There are a very few neurophysiological studies in which absence of sural SNAP and, normal median, ulnar, and peroneal motor conduction studies have been reported (Winkelman et al., 1962; Fig. 5.28).

The patients with Cockayne syndrome present with photosensitivity, deafness, microcephaly, and mental retardation. Patients with peripheral nerve involvement may have hyporeflexia and slowed motor conduction velocity. Sural nerve biopsy reveals segmental demyelination (Mossa and Dubowitz, 1970).

Cerebrotendinous xanthomatosis is a rare autosomal recessive lipid storage disease presenting with progressive dementia, ataxia, spasticity, tuberous cutaneous xanthomas, cataract, premature atherosclerosis, and peripheral neuropathy. The neuropathy in cerebrotendinous xanthomatosis is predominantly sensory due to loss of large myelinated fibers. Nerve conduction velocity is moderately slowed (Kuritsky et al., 1979).

Tangier disease is a rare disorder of plasma lipid transport. It results in profound deficiency of plasma high-density lipoprotein and widespread tissue storage of cholesteryl esters. It manifests with sensorimotor or mixed neuropathy. The patients complain of paresthesia, dysesthesia, hyperhidrosis, diplopia, and absent tendon reflexes. Three types of peripheral neuropathies have been described in Tangier disease:

Friedreich ataxia is an autosomal recessive hereditary ataxia affecting both ascending and descending tracts of spinal cord, degeneration of peripheral axons and myelin sheath. The genetic defect has been located to 9q13–21.1. The mutation of frataxin gene is an unstable expansion of the GAA trinucleotide repeat. The symptoms begin in legs in late childhood as progressive ataxia; clumsiness, and tremors appear later in hands with dysarthria and scanning speech. Examination reveals nystagmus, kyphoscoliosis, pes cavus, cocking of toes. The tendon reflexes are unelicitable and the plantar response extensor. Joint position and vibration sensations are absent in distal lower limbs. Pain and temperature may also be impaired in some patients. Motor nerve conduction studies are not very informative but SNAPs may be reduced or unrecordable (Dyck and Lambert, 1968).

Familial amyloid polyneuropathy is a group of autosomal dominant polyneuropathies characterized by deposition of extracellular amyloid in peripheral nerves and other organs. More than 10 different proteins may be deposited in tissue as amyloid protein, out of these, three aberrant proteins, i.e. transthyretin, apolipoprotein A1 and gelsolin are found in the peripheral nerves. Familial amyloid polyneuropathies are classified in Table 5.28.

The most common sites of nerve involvement in GBS are not randomly distributed throughout the nerve. In a study on 15 patients with GBS, conduction block was maximal in the terminal segment distal to wrist and to a lesser extent conduction block and conduction slowing were disproportionately created across the elbow and in the axila to spinal root segment. The vulnerability of these regions is attributed to relative deficiency of blood nerve barrier (Brown and Snow, 1991). The electrodiagnostic abnormalities in GBS in most studies reveal the involvement of distal and proximal nerve segments; prolongation of distal motor latencies and F-wave latencies being more common than slowed conduction velocity and conduction block (Albers et al., 1985). Others, however, have suggested more widespread involvement with sequential areas of demyelination scattered along the course of motor nerve (Asbury and Cornblath, 1990). The early preservation of motor conduction velocity in most GBS patients seems to support former position since numerous demyelinated segments along the course of the nerve should cumulatively slow conduction velocity over the conventional 100–300mm nerve segments studied. There may as well be different electrophysiological subpopulation of GBS patients reflecting varying distribution of demyelinated nerve segments (Van der Meche and Meulstee, 1988). The proposed diagnostic criteria of GBS serve as a research tool for uniform inclusion of patients. In clinical setting, one is not restricted by rigid criteria in the evaluation of possible GBS patients (Weinberg, 1999).

In GBS, every effort should be made to identify the region of conduction block. The diagnostic yield can be increased by:

Examination of waveform of motor or sensory action potential can add useful information and may require changing the conventional sensitivity and sweep speed of machine settings to record CMAP and late responses.

GBS has clearly defined subgroups, which include AIDP, acute motor axonal neuropathy (AMAN) and acute motor and sensory axonal neuropathy (AMSAN). In Europe and America, the underlying pathology is AIDP (Albany et al., 1989; Griffin et al., 1995); in which secondary by stander axonal degeneration is associated with poor prognosis (Van der Meche, 1991). It has been recognized that a primary axonal neuropathy is commoner in China and causes annual epidemics in children (McKhan et al., 1991) and may also be common in India (Gupta et al., 1994). In Chinese population, there is no difference in the speed of recovery between the neurophysiologically defined axonal and demyelinating subtypes (Ho et al., 1997).

Miller Fisher syndrome (MFS) is a variant of GBS comprising of ophthalmoplegia, ataxia, and areflexia (Fisher, 1956). It can follow campylobacter infection. IgG antibodies against ganglioside GQ1b are characteristic of this syndrome (Wilson et al., 1993). These antibodies recognize similar epitopes on C. jejuni and similar observation apply to GMI antibodies (Yuki et al., 1992). There is paucity of electrophysiological studies in MFS, that too on small number of patients. The striking feature was absence or low amplitude of SNAP and absence of H reflex with normal motor conduction velocity. These findings were regarded to be consistent with demyelinating peripheral neuropathy as seen in typical GBS (Guiloff, 1977). The H reflex studies in small muscles of hand and feet were used to study conduction in afferent fibers in a patient with MFS. The study revealed abnormalities, which correlated with degree of ataxia that contrasted with normal cutaneous sensory and motor functions and F-wave findings. This study suggested that ataxia in MFS may be due to demyelination in 1a afferent fibers (Weiss and White, 1986). Motor and sensory conduction and EMG in upper and lower limbs in 10 patients and cranial nerve conduction in seven patients were carried out. Electrophysiological abnormalities are consistent with axonal neuropathy or neuronopathy with predominant sensory nerve changes in the limbs and motor changes in cranial nerves. This feature was considered distinct from usual typical GBS (Fross and Daube, 1987). In a study on three patients with MFS, sensory motor conductions were within normal range but serial study revealed improvement in respective NCS parameters suggesting abnormalities in the initial record. H reflex elicited by stimulation of tibial nerve were absent in two, which became normally recordable at 6 months. F-waves recorded in upper and lower limbs were markedly prolonged out of proportion to their NCV slowing suggesting more proximal conduction abnormalities at the onset of illness with subsequent improvement. This study provided evidence of peripheral nerve dysfunction. On the other hand, there was no evidence of CNS abnormality by visual, ABR, and somatosensory evoked potentials as these were normal. The authors suggested that MFS should be included in the spectrum of GBS (Jamal and Ballentyne, 1988).

Neurophysiological changes in CIDP are similar to AIDP in the early stage and the differentiation between them is possible on the basis of clinical profile. In advanced CIDP, axonal degeneration may be superimposed on the features of demyelinating neuropathy, making the diagnosis more difficult. In CIDP, the motor nerve conduction velocity is below 80% of lower limit of normal value but may be less abnormal when the lesions are in spinal root and in the early stage. The amplitude of CMAP is also reduced because of dispersion or loss of motor units. Sensory conduction velocity may be slowed and the SNAP amplitude may be reduced or even be unrecordable. The neurophysiological criteria as per Ad Hoc Subcommittee of American Academy of Neurology AIDS Task Force are summarized in Table 5.32 (see also Fig. 5.29).

In CIDP and AIDP, concurrent central abnormalities have also been reported on MRI and evoked potential studies (Mendell et al., 1987). Eighteen patients with CIDP were subjected to visual and brainstem evoked potentials and MRI studies. The evoked potentials were abnormal in 9 out of 18 patients and 5 of them had central nervous system evidences of demyelination on MRI. Visual evoked potentials were suggestive of probable anterior optic pathway involvement in four patients, which was not detected by MRI (Pakalins et al., 1988). This raises the question of chance association of diseases of central and peripheral demyelination representing the spectrum of a single demyelinating disorder.

A number of medical conditions with peripheral neuropathy may fulfill the clinical criteria of CIDP such as HIV neuropathy, inflammatory bowel disease, systemic lupus erythematosus, diabetes mellitus, monoclonal gammopathy, POEM syndrome, lymphoma, Castleman disease, host versus graft disease, and drugs (cyclosporin and tacrolimus). The persistently distal weakness should prompt a search for HMSN1, Refsum disease, metachromatic leucodystrophy, and paraproteinemic neuropathy. Pain and sensory abnormality although occur in 10–15% patients with CIDP, if pain is out of proportion to weakness monoclonal gammopathy, Sjögren syndrome, and paraneoplastic neuropathy should be considered. Recently, CIDP-like clinical picture has also been reported in patients with mitochondrial neurogastrointestinal encephaloneuropathy (MINGE). These patients have clinical and electrodiagnostic features of CIDP and may even be treated for the same. In young patients with CIDP-like illness, absence of proximal weakness, presence of characteristic gastrointestinal features, lactic acidosis, extraocular abnormalities, CNS involvement, and lack of response to immunotherapy should suggest this possibility (Bedlack et al., 2004).

CIDP may be a more heterogenous condition than is generally appreciated. Majority of CIDP is sensorymotor (80%); however, 10% may be of pure motor and about 5–10% of pure sensory form. The pure sensory form although manifests with entirely sensory symptoms and signs, multifocal slowing of motor conduction is present. The disease course and response to steroids are similar to conventional sensory motor type of CIDP (Oh et al., 1992). Multifocal acquired demyelinating sensory and motor (MADSAM) neuropathy or Lewis Sumner variant showing multifocal conduction blocks with slowing of sensory motor conduction in the affected segments has also been described. It is controversial whether these are separate clinical entities or variants of CIDP.

The clinical, neurophysiological, CSF, and histopathological features simulating CIDP have been reported in diabetic patients. It may be difficult to diagnose CIDP on the basis of clinical ground alone. Some authors although highlighted predominant motor weakness (Uncini et al., 1999), others noted frequent complaints of imbalance (Gorson et al., 2000). In a recent study, however, no significant difference in clinical and neurophysiological findings were noted except older age of diabetic CIDP patients compared to nondiabetic CIDP (Haq et al., 2003). Recognition of CIDP in diabetics is important because it is amenable to IVIg or corticosteroid or both.

The patients with MMN are generally men and twothird are below 45 years of age. The weakness typically begins in one hand and may remain restricted to that hand for years or may even spread to involve all four limbs. There is remarkable focality of weakness and atrophy, which can be traced to individual peripheral nerve, often median and ulnar. Atrophy though is severe, in some muscles, the bulk may be preserved in spite of severe weakness unlike amyotrophic lateral sclerosis. Tendon reflex is reduced or absent in the affected limb; however, it may sometimes be normal or even brisk but is never associated with spasticity, pseudobulbar effect, or extensor plantar response (Pestronk et al., 1990). Cranial nerve involvement is rare and is seldom seen in the early stage of disease. Fasciculations and cramps are common and sensory examination normal. Clinically, MMN may simulate motor neuron disease but can be differentiated by the absence of cranial nerve palsy, absence of upper motor neuron signs and its slowly progressive course over years or decades and relative preservation of muscle bulk. The salient differentiating features between MMN and ALS are presented in Table 5.33.

The essential electrodiagnostic feature of MMN is persistent focal conduction block in one or more motor nerves localized in the areas not prone to compression. The commonest site of block is a distal forearm but may affect proximal segments even up to nerve root. The conduction block may be restricted to 3cm or may be extensive. More than one site in a single nerve may be involved. By the time the patient reports for the study, the conduction block is relatively severe; at least in some nerves with a drop of CMAP amplitude more than 80% or even complete block (Krarup et al., 1990). In addition to conduction block, there is dispersion of proximal motor response but it is not sufficient to account for the observed changes in CMAP amplitude. The nerve conduction velocity of the blocked segment is markedly slowed; however, the conduction velocity of the remaining nerve segment is normal, possibly due to short length of conduction block. The nerve conduction abnormalities are more widespread and there may be conduction slowing, temporal dispersion, prolonged distal motor latency, and prolonged F latency in the nerves without conduction block (Katz et al., 1997). Sensory conduction is normal including the segment showing conduction block. Needle EMG is always abnormal in MMN with prominent fibrillations in the most severely involved muscles. Fasciculations are common and grouped repetitive discharges (myokymia) are also seen. Myokymia is an important finding of demyelinating neuropathy and is rare in ALS. The problems in the diagnosis of MMN are attributed to its initial description from group of CIDP and ALS patients. Different authors have used a broad range of diagnostic criteria. Consensus criteria for the diagnosis of MMN have been provided by the American Academy of Electrodiagnostic Medicine (AAEM) and are summarized in Table 5.34 (Richard et al., 2003).

Demyelinating neuropathies have been reported in 30–33% of HIV seropositive patients referred for the evaluation of neuropathy (Cornblath et al., 1988; Miller et al., 1988a). The clinical picture of the patients with demyelinating neuropathy in association with HIV infection is similar to AIDP or CIDP. There is no difference in age of onset; sex or prodromal infection with HIV-associated GB syndrome and those without HIV infection. There is also no difference in the time to reach maximum weakness and time taken for improvement and frequency of cranial nerve palsy in both the groups (Thorton et al., 1991). The patients with AIDP have acute progressive weakness in distal and proximal muscles, which is associated with areflexia and mild sensory impairment. The neurophysiological signs of acquired demyelination are generally more prominent in CIDP compared to AIDP. The neurophysiological abnormalities include slowed motor nerve conduction velocity, conduction block, prolonged F-wave latency, and unrecordable or slowing of sensory conduction. There may be reduced or absent SNAP (Cornblath, 1988; Miller et al., 1988b). On EMG, there may be reduction of MUP recruitment proportionate to the degree of weakness. Varying amount of fibrillations may also be present. The most important abnormality in inflammatory demyelinating neuropathy associated with HIV is in CSF. The CSF protein is elevated from 50–250mg/dl, and lymphocytic pleocytosis of 20–50/mm³ is seen in majority of patients (Ebright and Crane, 1991). In a study on AIDP patients with HIV-I infection, the mean CSF cell count was 14/mm³ and 1/mm³ in those not associated with HIV-I (Thorton et al., 1991). Presence of pleocytosis in a patient with AIDP or CIDP should alert for the possibility of coexistent HIV infection.

Distal symmetrical axonal neuropathy manifesting with painful dyesthesias is the commonest neuropathy that occurs in patients with HIV infection. It has been reported in one-third patients with AIDS in whom other recognized causes of polyneuropathy have been excluded. This type of polyneuropathy is detected pathologically in almost all the patients dying of AIDS (Griffin et al., 1991; So et al., 1988). The chief symptoms are burning in feet, painful soles which are worsened by touch or pressure and numbness. The feet are affected more than the hands. Tendon reflexes may be absent. An elevated threshold to vibration, pinprick, and thermal sensations may be present in 85–100% patients with AIDS (Cornblath and McArthur, 1988; So et al., 1988). Mild weakness of toe movements and wasting of intrinsic foot muscles are common. Most AIDS patients who develop distal sensory polyneuropathy also have constitutional symptoms and weight loss. In distal symmetrical polyneuropathy, SNAPs are small or unrecordable. Less commonly sensory or motor amplitude of median or ulnar nerves are also reduced although nerve conduction velocities are normal or only mildly reduced (Cornblath and McArthur, 1988; Snider et al., 1983). Electromyography may demonstrate acute or chronic partial denervation and reinnervation in distal leg muscles (Cornblath and McArthur, 1988). Histopathological picture of sural nerve biopsy reveals axonal degeneration of myelinated and unmyelinated fibers (Bailey et al., 1988). Some authors have noted demyelination but it does not appear to be segmental. Epineural and endoneural perivascular infiltration has been reported in some patients (Mah et al., 1988). Distal axonopathy in HIV-I patients is attributed to distal degeneration of sensory and motor axons. Such a clinical syndrome may be caused by nutritional deficiency, alcohol, drug toxicity, diabetes mellitus, and syphilis. The cause of axonal neuropathy in HIV patients is unknown. A subset of HIV positive patients with distal symmetrical polyneuropathy have been found to have necrotizing vasculitis (Bradley and Verma, 1996).

The patients with multiple mononeuropathy present with multifocal sensorimotor complaints in the distribution of cutaneous nerves, mixed nerves, and nerve roots. In nine homosexual men with lymphadenopathy and multifocal mononeuropathy, five had cranial nerve involvement and five improved either spontaneously or concurrently with plasmapheresis (Lipkin et al., 1985). There are probably two forms of multiple mononeuropathies in association with HIV infection: (i) limited distribution of multiple mononeuropathy presenting with acute onset of sensory motor deficit involving one or two peripheral nerves possibly including facial nerve. These patients may have AIDS-related complex or previously asymptomatic HIV infection, which is associated with CD4 count >200. Additional neuropathies usually do not develop and the neuropathy improves over several months with or without immunosuppressive treatment (Lipkin et al., 1985). (ii) The extensive form of multiple mononeuropathy presents with the involvement of three or more nerves in two or more limbs, sometimes including laryngeal nerve. In these patients, CD4 count is <50 and cytomegalo virus infection has often been found. In the absence of CMV infection, necrotizing vasculitis should be suspected. The neurophysiological findings include reduction in amplitude of SNAP and CMAP and mild reduction of nerve conduction velocity. The EMG reveals denervation and neurogenic MUPs. The presence of asymmetric or focal axonal lesion is helpful in the diagnosis of this type of neuropathy (Lange et al., 1988). The electrodiagnostic findings may, however, be diffuse and symmetrical similar to the distal symmetric neuropathies (Gheradi et al., 1989) or may suggest demyelination (Lipkin, 1985). Histopathological changes in multifocal neuropathies include axonal degeneration in the early stage, polymorphonuclear infiltrates, and mixed axonal and demyelinating lesions and less commonly necrotizing arteritis.

Progressive polyradiculopathy occurs late in the course of HIV disease when patients generally have advanced immunosuppression and low CD4 counts. The patients complain of lower extremity and sacral paresthesia followed by rapidly progressive paraparesis, areflexia, ascending sensory loss, and urinary retention. Pain in the distribution of cauda equina is common. On examination, usually there is minor sensory loss, lower limb areflexia, and a thoracic sensory level has occasionally been reported (de Gans et al., 1990). The other signs of pyramidal dysfunction such as hyperreflexia, spasticity, and extensor plantar are generally absent. Upper limbs remain normal till late in the course of polyradiculopathy. Cerebrospinal fluid reveals polymorphonuclear pleocytosis, elevated protein, and low glucose. Cytomegalo virus (CMV) may be isolated from CSF. On EMG, there is widespread denervation in the lower limb muscles characterized by fibrillations, sharp waves and complex repetitive discharges and neurogenic MUPs with reduced recruitment (Dalakas and Pezesckpour, 1988). Sensory and motor conductions are generally normal or only slightly affected, but F-waves may be delayed or unrecordable from the affected muscles (So et al., 1990). The amplitude of CMAP declines as the disease progresses. These neurophysiological findings are consistent with proximal axonal pathology in lumbar roots. The predominant histopathological features are marked inflammation and extensive necrosis of ventral and dorsal nerve roots (de Gans et al., 1990). The inflammation is most prominent in the lumbar region and less severe in caudal and rostral direction. Cranial nerves and spinal cord may be involved close to inflamed roots (Miller et al., 1990). In almost all the autopsied patients with polyradiculopathy, the histological picture is consistent with CMV infection; therefore, polyradiculopathy in AIDS is attributed to CMV infection. Isolation of CMV from CSF and therapeutic response to gancylovir further support the role of CMV in polyradiculoneuropathy. The other rare causes of lumbosacral polyradiculoneuropathy in AIDS include herpes zoster, syphilis, mycobacterial infection, toxoplasmosis, and leptomeningeal lymphomatosis (Gherardi et al., 1998).

The prevalence of autonomic neuropathy in AIDS is not known but orthostatic hypotension, syncope, impotence, urinary dysfunction, diminished sweating, diarrhea, and cardiac arrhythmias have been reported. In a study on 25 HIV infected patients, three had orthostatic hypotension and dizziness (Evenhouse et al., 1987). The patients with advanced disease have a higher risk of developing autonomic neuropathy. Both central and peripheral nervous system abnormalities may contribute to autonomic symptoms in AIDS patients. In an autopsy study on six patients with AIDS, abnormalities were present in all the patients in cervical sympathetic ganglia, and five had diarrhea; but autonomic functions were not evaluated in these patients (Chimelli and Scaravilli, 1991). Besides these major types of neuropathies, associated vasculitis, malignancy and drug toxicity also result in peripheral neuropathy in AIDS patients. Table 5.36 summarizes various types of neuropathies in HIV patients.

Three cardinal signs remain the mainstay for the diagnosis of leprosy in clinical practice—anesthetic skin lesions, enlarged peripheral nerves, and AFB in skin smear or biopsy. Disease onset is insidious; skin and peripheral nerves are the primary targets. The cutaneous lesions in leprosy are present in more than half the patients and may manifest with area of sensory loss, anhydrosis, hair loss, and atrophic cutaneous patches with painful enlargement of nerve trunks. Plantar ulcer and other trophic changes occur as late sequela. Sensory loss is the commonest finding of leprous neuropathy and is due to mixed dermal nerve and nerve trunk damage. The sensory loss in leprosy is extremely variable ranging from a small hypoesthetic patch to severe sensory loss over most of the body surface except body folds. Early skin lesions show impairment of light touch, loss of thermal, and pain sensations. Proprioception is preserved; therefore, the patient can use largely anesthetic limbs efficiently which results in trauma, trophic changes, and recurrent infections leading to further deformities. Loss of pigment in the affected nerve territory results in depigmented anesthetic patches, which are characteristic especially in dark skinned patients. Cooler areas of the body are most vulnerable to leprous neuropathy (Hastings et al., 1968). In some patients, the affected area may have dissociated sensory loss, i.e., loss of pain and temperature with preservation of touch. The sensory loss in leprosy ranges from the area of sensory nerve ending and to limited number of nerve fascicles of a nerve trunk or even large nerve trunk. Among the large nerve trunk, most commonly affected are ulnar, common peroneal, median, posterior tibial, superficial radial, greater auricular (Fig. 5.30), and facial nerves (Dastur, 1955). Nerve trunks are probably enlarged in about onethird patients with leprosy (Said, 1980). Superficial nerve such as greater auricular, supraorbital branch of trigeminal, or large nerve trunk especially ulnar nerve above elbow, peroneal and radial cutaneous nerves are thickened. Nerve hypertrophy occasionally may be associated with spontaneous tingling or pain. The other causes of hypertrophy of nerve are hypertrophic neuritis, neuroma, neurofibromatosis, and amyloidosis. In presence of isolated nerve hypertrophy the diagnosis of leprosy should be made carefully.

The neuropathy in lepromatous leprosy is relatively slow and progressive compared to other varieties. It is characterized by more widespread involvement of skin and nerves causing bilateral symmetrical distal polyneuropathy (Fig. 5.31). In the early stage sensory loss may appear in the pinna of the ear, dorsum of hand, dorsomedial forearm, dorsum of feet, and anterior calves. With progression of disease, the sensory loss spreads to malar area, nose, breast, abdomen, and buttocks. In the later stage there may be weakness. The earliest motor deficit is seen in ulnar innervated intrinsic hand muscles leading to clawing of hands. Characteristically, palms and soles are spared in spite of extensive sensory loss in other parts of the body. There is sparing of scalp “Hairline sign” due to increased temperature of scalp (Dutta et al., 1983). Skin lesions in lepromatous leprosy are numerous and include macules, papules, and nodules with infiltration and thickening of skin. In these patients AFB may be found in skin and nasal smear. The more lepromatous the findings, less marked are the symptoms (Pedley et al., 1980).

Pure neural leprosy refers to neurologic deficit and nerve thickening with or without tenderness in the absence of skin involvement. It has been reported in 10% patients attending to a leprosy center in India (Uplekar and Antia, 1986). Most of these patients present with the clinical picture simulating mononeuritis multiplex. Electrodiagnostic studies reveal demyelinating features in the early and axonal in the late stage. This group of leprosy patients is difficult to diagnose as nerve biopsy demonstrates epithelioid granulomas in only 14% cases and AFB is seen in 16% cases; however, PCR was positive in 47% (Jardin et al., 2003). Absence of skin lesions and nondiagnostic nerve biopsy often leads to delay in diagnosis of pure neural leprosy.

Electrodiagnostic studies are useful in confirming the segmental nature of Hansen’s disease, estimating the extent of involvement and serve as an objective index of response to treatment. Motor nerve conduction studies have been carried out in lepromatous, borderline, and tuberculoid leprosy and have confirmed the segmental slowing in nerve trunk and branches at the site of enlargement and entrapment. Maximum slowing of nerve conduction occurred in 16 of 27 ulnar nerves at the elbow and 6 of 27 median nerves just proximal to the carpal tunnel in the lower third of forearm (Divekar, 1965). In a study on lepromatous leprosy, the slowing of conduction in the elbow segment of ulnar nerve was reported (Hackett et al., 1968), whereas conduction at wrist, axilla to mid humerus and forearm segments were spared. In the lower extremities peroneal conduction studies usually reveal slowing of nerve conduction across the fibular head and distal latency prolongation at the ankle (Swift et al., 1973). In a patient with lepromatous leprosy, slowing of motor nerve conduction of ulnar, median, peroneal, and tibial nerves have been reported (Rosenberg and Lovelace, 1969). Nerve conduction velocities generally correlate with clinical weakness and nerve enlargement (NdiayeNiang et al., 1986; Verghese et al., 1970). In leprosy, the nerve conduction slowing has also been reported without any clinical evidence of neuropathy (McLeod et al., 1975). Of 71 ulnar and median nerve conduction studies in leprosy patients without any muscle weakness, nerve conduction velocity was normal in 69. Ulnar nerve conduction velocity was slow in the upper arm including the elbow segment compared to forearm. Median nerve conduction slowing occurred in distal third of forearm in three patients (41m/s, 12.2m/s and unrecordable) rather than in carpal tunnel (Verghese et al., 1970). Facial nerve conduction in leprosy has been extensively studied. Patchy and segmental involvement of the facial nerve has been reported and the zygomatic branch was found to be more frequently affected compared to the other branches (Fig. 5.32; Monrad and Krohn, 1923). In another study, orbicularis oculi was more frequently involved compared to orbicularis oris in leprosy patients. This observation was substantiated clinically, electrophysiologically and histopathologically (Dastur et al., 1968). In patients with lagophthalmos, the facial nerve latency to orbicularis oculi was most frequently prolonged; whereas in some patients the latency to frontalis and orbicularis oris was also prolonged (Divekar, 1965; Chaco et al., 1968). In leprosy patients with sensory loss, the SNAPs are unrecordable (Divekar, 1965). The lepromatous leprosy patients have normal sensory nerve conduction velocity, but the amplitude of SNAP may be reduced (Rosenberg and Lovelace, 1969). Radial nerve sensory conduction abnormality has been reported to be a sensitive indicator of neural abnormality in leprosy (Sebille, 1978). Sensory conduction in the lower extremity is more frequently abnormal compared to the upper (Ndiaye-Niang et al., 1986). Electromyography in leprosy patients commonly reveals neurogenic changes, however, occasionally short duration polyphasic low amplitude potentials suggestive of myopathy have been reported (Divekar, 1965; Magora et al., 1965) although such EMG changes may be found in acute denervation as well.

Diabetic neuropathy (DN) refers to symptoms and signs of neuropathy in a diabetic in whom other causes have been excluded. Diabetic neuropathy is one of the commonest causes of neuropathy. It is probably not a single pathologic entity, but a syndrome of different clinical presentations due to diverse etiologies such as metabolic, ischemic, or both. Distal symmetrical polyneuropathy although is the commonest but a number of peripheral neuropathy syndromes are associated with diabetes, which are summarized in Table 5.37.

For diagnosis of DN, bedside examination should include assessment of muscle power, sensation of pinprick, joint position, touch and temperature. Vibration test should be carried out by tunning fork of 128Hz. For touch sensation, monofilament of 1g is recommended. Sensory examination should be performed on hands and feet bilaterally. In old age (>70 years) vibration and ankle reflex may be reduced normally and considered abnormal if these are absent rather than reduced in a patient with DN. Quantitative sensory testing may be used as ancillary test but is not recommended for routine clinical practice (Shy et al., 2003). The autonomic function tests commonly used in DM are based on BP and heart rate response to a series of maneuvers. Specific tests are used in evaluating gastrointestinal, genitourinary, sudomotor function, and peripheral skin blood flow. Nerve biopsy may be useful for excluding other causes of neuropathy. Skin biopsy has been used in the diagnosis of small fiber neuropathy by quantification of PGP 9.5 when all other measures are negative (Kennedy et al., 1996). Diabetes as a cause of neuropathy is diagnosed by exclusion of other causes in patients who present with painful feet and have impaired GTT (Sumner et al., 2003). Recently, confocal corneal microscopy in the assessment of diabetic polyneuropathy has been reported. This noninvasive technique may have great potential in assessing nerve structure in vivo without need for nerve biopsy (Quattrini et al., 2003).

Motor nerve conduction, F response and sensory nerve conduction studies are important methods of documentation and follow up of nerve functions in diabetic neuropathy. Motor nerve conduction studies are affected in a small subset of diabetic neuropathy (large fiber neuropathies). Even in large diameter fiber neuropathy NCV is insensitive for many pathological changes known to be associated with DN. The nerve conduction changes are nonspecific and key to the diagnosis lies in excluding other causes or those superimposed on DN. Entrapment neuropathies are common in diabetics and result in unilateral NCV changes, especially across the entrapped segment of the nerve. The commonest abnormality in diabetics is reduction in the amplitude of motor or sensory action potential because of axonopathy. Marked slowing of NCV suggests demyelinating neuropathy, which is rarely associated with diabetes; therefore marked slowing of NCV in a diabetic should prompt investigations for an alternative diagnosis. However, the likelihood of CIDP occurring in a diabetic is 11 times higher than in normal population (Sharma et al., 2002). The NCV is gradually diminished by DN, with estimate of a loss of approximately 0.5m/s/year (Arezzo, 1997).

In renal failure, polyneuropathy is common. Nearly, all the patients requiring dialysis have evidence of distal symmetrical sensorimotor polyneuropathy. In many patients, it manifests with mixed sensorimotor polyneuropathy with borderline reduction of motor and sensory evoked amplitudes (Bolton, 1980). In other patients pronounced slowing of nerve conduction velocity with reduced proximal CMAP is present suggesting segmental demyelination superimposed on axonal loss, which has been verified pathologically (Dyck et al., 1971; Roy et al., in press). On EMG, fibrillations are seen in the distal muscles particularly in foot. Femoral neuropathy following renal transplant surgery, and femoral and radial neuropathy following vascular access for hemodialysis in patients with renal failure have also been reported (Kumar et al., 1991; Roy et al., 1998; Kalita et al., 1995).

Acute intermittent porphyria is an autosomal dominant disorder of porphyrin metabolism with incomplete penetrance. It presents with characteristic triad of abdominal cramps, psychosis, and polyneuropathy. The exact incidence of peripheral neuropathy in porphyria is not known but is estimated to be 10–40% based on the information derived from patients diagnosed to have porphyria (Albers et al., 1978; Sorensen and With, 1971). The neurologic findings resemble AIDP in the early stage. Weakness usually develops 2 or 3 days after the onset of abdominal and psychiatric symptoms. In one study, 80% of the patients developed neuropathy within 1 month of abdominal pain (Ridley, 1969). Once neuropathy develops it may progress to the maximum deficit over one month (Sack, 1990). Weakness may begin in upper limbs and cranial nerves innervated muscles (Fig. 5.33). The proximal and distal muscles are equally affected. Unlike AIDP, the paralysis is not ascending. In a study about half the patients with porphyric neuropathy experienced onset in the arms and 80% had more marked proximal weakness than the distal. One-third patients had lower limb onset, of whom 50% had more marked proximal weakness whereas remaining patients had generalized weakness. Asymmetric weakness is a common finding (Ridley, 1969). Muscle atrophy is seen early in the course of the disease. Though several forms of porphyric neuropathy have been reported, however, typical neuropathy is considered to be a primary motor neuropathy (Leger and Salachas, 2001). Motor neuropathy is characterized by weakness, areflexia, and sensory symptoms with or without demonstrable sensory loss. Sensory loss is variable and may be either symmetric glove or stocking type or involve the upper limb and trunk (bathing suit distribution), which may be patchy and migratory. Hyporeflexia in porphyria is proportional to the degree of weakness, whereas in AIDP the reflex is lost early.

Primary systemic amyloidosis rarely results in subacute or chronic form of sensorimotor neuropathy. Sometimes multiple myeloma and macroglobulinemia also result in neuropathy because of amyloidosis. About one-sixth of patients with primary amyloidosis may have neuropathy. The patients present with distal symmetrical sensory loss; but the most striking feature is pronounced dysautonomia, which results in postural hypotension and bowel bladder involvement (Misra et al., 1994). In a study on 18 immunohistochemically confirmed patients with primary amyloidosis, 14 had loss of feeling, paresthesia, and pain; and four primarily presented with weakness, weight loss, and postural hypotension. Typically all developed sensory, motor, and autonomic symptoms over time. The neuropathy in primary amyloidosis is similar to that in familial amyloidosis (Kyle and Dyck, 1993). Familial amyloid neuropathy has been described earlier.

Thiamine, pyridoxine, and vitamin B₁₂ deficiencies are most commonly associated with nutritional neuropathy. Alcohol-related neuropathy is also linked to thiamine deficiency although it possibly has direct toxic effect on the nerve. Thiamine deficiency is responsible for beriberi, which results in heart failure (wet beriberi) and neuropathy (dry beriberi). Beriberi has been reported in the individuals who consume rice as a dietary staple, especially the polished rice in which the pericarp rich in thiamine has been removed. The excessive use of alcohol as a calorie source is the probable basis of thiamine deficiency in alcoholism. Patients with thiamine deficiency neuropathy, present with burning feet syndrome followed by distal symmetrical weakness and wasting. Skin is glossy, atrophic, and hairless due to autonomic dysfunction. Rarely tongue, facial, and laryngeal weakness have also been reported. Pathologically, there is Wallerian or axonal degeneration in the distal nerve segments (Windenbank, 1993). Vitamin B₁₂ deficiency most commonly results from vitamin B₁₂ malabsorption due to pernicious anemia although it can occur with diphyllobothrium latum infestation, strict vegetarianism, and steatorrhea (Britt et al., 1971; Badenoch et al., 1955; Misra et al., 2003). In vitamin B₁₂ deficiency, peripheral neuropathy is rarer (5%) compared to spinal cord involvement (80%). The common neurologic signs in B₁₂ deficiency include hyporeflexia or areflexia at the ankle with plantar extensor. The prominent sensory symptoms are due to posterior column involvement (Mayer, 1965). Electrophysiological features include slowing of nerve conduction, reduced amplitude, or unrecordable SNAP (Mayer, 1965; Gilliatt, 1966). Pyridoxine deficiency resulting in peripheral neuropathy is encountered in patients undergoing isonicotinic acid hydrazide and hydralazine therapy. The patients complain of distal symmetrical numbness and tingling, which can be reversed by discontinuation of drug (Cavalier and Gambetti, 1981; Selikoff et al., 1952).

A rare neurologic disorder of childhood consisting of spinocerebellar degeneration in association with peripheral neuropathy and pigmentary retinopathy has been related to vitamin E deficiency after prolonged intestinal malabsorption (Sokol et al., 1988). The same mechanism has been proposed to explain the neurologic disorders with abetalipoproteinemia, fibrocystic disease, and extensive intestinal resection (Harding, 1985). In these patients ataxia, loss of tendon reflexes, ophthalmoparesis, proximal muscle weakness, elevated serum CK, and decreased sensations are the usual manifestations. The neurologic functions improve after long-term correction of vitamin E deficiency. The sensory nerve conduction is either not recordable or the amplitude of SNAP is reduced. Motor nerve conduction studies are normal, however, in the later stage, the CMAP may be slightly reduced due to disuse atrophy, axonal stenosis or combination of both. Needle EMG is normal except in the later stage when fibrillations and sharp waves may be recorded in the distal muscles.

Electrophysiological findings in CIP are typical of primary axonal degeneration of motor and sensory fibers in multiple nerves (Bolton, 1995). Since reduced CMAP, fibrillations and positive sharp waves are also seen in primary myopathies, CIP can be diagnosed with certainty only if SNAPs are reduced. In the presence of limb edema SNAP recording may be flawed and may necessitate near nerve needle recording or sequential studies. Phrenic nerve conduction and needle EMG of diaphragm should be performed to establish if CIP is the cause of difficulty in weaning from ventilator (Bolton, 1993).

Vasculitic neuropathy is due to infarction of one or more nerves produced by inflammatory occlusion of blood vessels (Albert et al., 1988). The pathological hallmark of vasculitis is segmental fibrinoid necrosis of a vessel wall and transmural inflammatory cell infiltration (Bouche et al., 1986). The underlying mechanism of vasculitic neuropathy is nerve infarction due to occlusion of vasa nervosa of 75–200μm diameter epineural arteries. This results in axonal degeneration with a central fascicular distribution at the level of infarction (Dyck et al., 1972). Sometimes demyelination and conduction block may also occur but are transient and not prominent (Perry et al., 1981; Dyck et al., 1972). In experimental studies, the large myelinated fibers have been found to be less vulnerable to ischemia compared to small fibers (Perry and Brown, 1982); but this has not been consistently found in clinical studies on polyarteritis nodosa and rheumatoid arthritis (Conn et al., 1972; Vital and Vital, 1985).

This group comprises polyarteritis nodosa (PAN), Churg–Strauss syndrome, and an overlap syndrome. The diagnosis rests on demonstrating necrotizing vasculitis at least in two organ systems. In this group, peripheral neuropathy occurs in 44–60% patients (Cohen et al., 1980; Frohnert and Sheps, 1967; Travers et al., 1979). The presence of vasculitic neuropathy itself satisfies the criteria of involvement of one organ system. Most of the neuropathies in patients with PAN are vasculitic in nature and comprise mononeuropathy multiplex or asymmetrical polyneuropathy (Bouche et al., 1986). Neuropathy is the major presenting feature in more than one-third patients with PAN, which usually develops in the first year of illness. Patients generally have the signs and symptoms of systemic disease and laboratory evidences of multiple organ involvement. Weight loss, fever (66–76%), cutaneous sign and arthralgia (44–58%), hypertension (28–34%), renal failure, and gastrointestinal involvement (8–31%) have been reported (Frohnert and Sheps, 1967; Guillevin et al., 1988). The abnormalities in laboratory findings include raised ESR, leucocytosis; anemia, abnormal urinalysis, and presence of HBsAg. The diagnosis of PAN group of disorders requires histopathologic documentation of necrotizing vasculitis or angiographic documentation of aneurysm at the bifurcation or segmental narrowing in renal, hepatic, or mesenteric vessels. Angiography has been reported to suggest the diagnosis of PAN in 60% cases (Albert et al., 1988; Travers et al., 1979). The yield of nerve and muscle biopsy is higher especially if directed at the site of electrodiagnostic abnormalities (Albert et al., 1988); therefore muscle nerve biopsy should be the initial investigation for confirming the diagnosis of PAN. Churg–Strauss syndrome is differentiated from PAN by the presence of asthma, peripheral eosinophilia, eosinophilic infiltrates, or granulomata. The peripheral nerve involvement is similar to PAN. There is controversy regarding independent status of PAN and Churg–Strauss syndrome (Guillevin et al., 1988; Moore and Cupps, 1983).

Rheumatoid arthritis is the commonest connective tissue disorder and systemic arteritis occurs in 8–24% patients of long standing rheumatoid arthritis for an average of 14 years (Conn et al., 1984; Sokoloff et al., 1951). Systemic arteritis is associated with weight loss, rheumatoid nodules, and cutaneous lesions in more than 80% patients. Of these patients vasculitic neuropathy is present in 40–50% (Cruickshank, 1954; Scott et al., 1981). Neuropathy in rheumatoid arthritis is clinically, pathologically, and electrophysiologically similar to that in PAN. Although two-third patients with rheumatoid arthritis may develop mild sensory neuropathy, clinically significant vasculitic neuropathy however develops only in 1–10% (Conn et al., 1984). In rheumatoid vasculitic neuropathy, ESR and rheumatoid factor titer are high and associated with a poor prognosis.

Distal symmetrical polyneuropathy manifests with subacute or chronic evolution of sensory symptoms, which are attributed to axonal degeneration. In a study on seven consecutive patients with SLE, the pathological evidence of vasculitis was subtle with perivascular mononuclear infiltrates and intimal thickening. In these patients immunological factors other than the vasculitis leading to axonal degeneration were suggested (McCombe et al., 1987). This type of neuropathy is common in rheumatoid arthritis occurring in up to 75% patients when detailed clinical and electrodiagnostic studies have been undertaken (Conn et al., 1984; Good et al., 1965). These histological changes were also found in autopsy of the patients with rheumatoid arthritis who did not have clinically apparent neuropathy (Conn et al., 1984). Peripheral neuropathy has been reported in 10–15% patients with Sjögren’s syndrome. The most frequent type of neuropathy is mixed distal symmetrical but pure sensory and asymmetrical pattern may also be present (Mellgren et al., 1989). In giant cell arteritis, peripheral neuropathy has been reported in 14% patients with about half having diffused peripheral neuropathy. In majority of patients the neuropathy improves with steroid treatment (Caselli et al., 1988).

The association between connective tissue disease and compression neuropathy has been most extensively studied for carpal tunnel syndrome. In a study, more than one-third patients, with rheumatoid arthritis had symptoms of carpal tunnel syndrome and a positive Tinnel’s sign. The symptomatic patients had significantly slower conduction through the carpal tunnel compared to asymptomatic patients (Herbison et al., 1973). An increased incidence of ulnar neuropathy at elbow and peroneal neuropathy at the fibular head has also been reported in rheumatoid arthritis (Balagtas Balmaseda et al., 1983; Nakano, 1975). Other connective tissue diseases reported to be associated with carpal tunnel syndrome are Sjögren’s syndrome in 20% patients, giant cell arteritis and SLE (Caselli et al., 1988; Molina et al., 1985; Sidiq et al., 1972).

Trigeminal sensory neuropathy is characterized by unilateral or bilateral facial numbness and is occasionally associated with pain. Trigeminal neuropathy has been reported with Sjögren’s syndrome, SLE, mixed connective tissue disorders, and dermatomyositis. Trigeminal neuralgia may be the presenting feature of progressive systemic sclerosis and mixed connective tissue disorders (Farrel and Medsger, 1982; Lecky et al., 1987). Blink reflex study has revealed an efferent delay (delayed ipsilateral R₁ and R₂) or an absent response in about half of the patients (Lecky et al., 1987). The available pathological data support degeneration of peripheral myelinated axons from a lesion distal to Gasserian ganglion. The cause of sensory neuropathy is attributed to either vasculitis or fibrosis (Lecky et al., 1987).

Sensory neuropathy has been increasingly recognized with Sjögren’s syndrome. The chief complaints are usually due to impaired joint position sensation resulting in gait ataxia, clumsiness, and pseudoathetosis. Its onset is acute or indolent. SNAPs are typically absent or small. Nerve conduction velocity is generally normal. Cutaneous nerve biopsy reveals predominant loss of large myelinated fibers with perivascular inflammatory cell infiltration without necrotizing vasculitis. Dorsal root ganglia specimens have also revealed lymphocytic infiltration (Griffin et al., 1990).