Molecular genetics and neuropathology of Charcot-Marie-Tooth disease type 1A

Charcot-Marie-Tooth (CMT) syndrome describes a genetically and clinically heterogeneous group of polyneuropathies. Electrophysiologically, at least two types of CMT can be distinguished; CMT1 which has decreased nerve conduction velocities (NCV) and CMT2 which has normal or near normal NCV with decreased amplitudes. For CMT1, three gene loci (on chromosomes 1, 17 and the X chromosome) have been mapped. The locus on chromosome 17, CMT type 1A (CMT1A), is responsible for the most common form of CMT which has recently been shown to be associated with a large DNA duplication. Recent data demonstrates that the CMT1A phenotype results from an inherited DNA rearrangement and a gene dosage effect. The trembler (Tr) and allelic tremblerJ (TrJ) mice have been proposed as animal models for CMT. Tr has similar electrophysiological and neuropathological features to CMT1 patients and maps to mouse chromosome 11 in a region of conserved synteny with human chromosome 17p. Tr and TrJ have recently been shown to have different point mutations in regions encoding putative transmembrane domains of the myelin specific protein PMP-22. The human peripheral nerve-specific PMP-22 gene maps within the CMT1A duplication. PMP-22 is thus a candidate gene for CMT1A. This paper describes the molecular genetics of CMT1A and sural nerve pathology in CMT1A patients with the CMT1A duplication.

[Mitochondrial myopathy with peripheral neuropathy]

This paper reported mitochondrial myopathy with peripheral neuropathy of 2 cases. Both patients were males. Age: 22, 32.

DURATION

11, 14 years respectively. They showed recurrent paralysis and asthenia of limbs. Case 1 was motor sensory neuropathy, whose EMG revealed neurogenic injury. Case 2 involved only the lower limbs, whose lactic acid level was increased. In both patients, muscle biopsy showed Ragged Red fibers and abnormal mitochondria. Sural nerve biopsy revealed moderate reduction in the number of mylinated fibers and chronic axonal degeneration without regeneration cluster and hypertrophic neuropathy. Electron microscopic examination showed the increase of Schwann cells and mitochondria, and abnormal mitochondria being less marded in Case 2. Mitochondrial myopathy with peripheral neuropathy and its pathogenesis were discussed.

[Krabbe's disease (globoid cell leukodystrophy). Apropos of 5 cases]

A clinical study on five cases of Krabbe's disease (globoid cell leukodystrophy) was performed. A final diagnosis was done either with post-mortem study (two cases) or by enzymatic assays carried on cultured fibroblasts (two cases). Peripheral nerve biopsy for electron microscopy was performed in all cases, and the ultrastructural alterations characteristics of Krabbe's disease were always found. The authors emphasize the suggestive clinical and laboratory data which enable the diagnosis of Krabbe's disease in the absence of the ultrastructural exam of peripheral nerve, or the enzymatic assays not performed in this country.

Subclinical diabetic neuropathy: similarities between electrophysiological results of patients with type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetes mellitus

Type 1 (insulin-dependent) and Type 2 (non-insulin-dependent) diabetic patients share many clinical and biochemical characteristics. However, sural nerve biopsies from patients with advanced and chronic neuropathy show ultrastructural differences between these two groups. We investigated whether at a subclinical stage of the illness, when Type 1 and Type 2 diabetic patients are clinically uniform and the histopathological nerve alterations are not advanced, comparison between the two diabetes groups might show differences in nerve fibre involvement related to the different pathogeneses of the neuropathies. A total of 88 diabetic patients (52 Type 1 and 36 Type 2), with a subclinical form of polyneuropathy were selected. The clinical neurophysiological examination consisted of motor and sensory nerve conduction studies, Hoffmann (H)-reflex, single fibre electromyography and static as well as dynamic pupillometry. With regard to clinical neurophysiological abnormalities, the severity of the polyneuropathy appeared to be equal in both groups. Despite the absence of clinical symptoms the neurophysiological abnormalities were pronounced and it was impossible to differentiate Type 1 diabetic patients from Type 2 diabetic patients on a clinical neurophysiology basis when correcting for differences in age, height, and duration of illness. In the Type 1 diabetic group as well as in the Type 2 diabetic group the autonomic nerve fibres and nerves in the legs were more frequently affected than the thick myelinated nerves in the arms. These findings do not support the assumption that there is a difference in the manifestation of polyneuropathy between Type 1 and Type 2 diabetic patients.

The reproducibility and sensitivity of sural nerve morphometry in the assessment of diabetic peripheral polyneuropathy

The nerve fibre loss, atrophy and injury of diabetic peripheral polyneuropathy and their responses to metabolic intervention have been studied by morphometric analysis of sural nerve biopsies. The magnitudes and sources of intra- and inter-individual variation in these morphometric measures have not been investigated previously in a systematic manner. Morphometric parameters of nerve fibre damage were measured in four separate fascicles from bilateral sural nerve specimens obtained post-mortem from 13 diabetic and 13 non-diabetic subjects. Intra- and inter-individual coefficients of variation were computed and compared to the magnitude of the differences between normal and diabetic subjects. Several morphometric variables emerged as highly sensitive and reproducible measures of nerve fibre damage suitable for clinical studies of diabetic peripheral polyneuropathy. These observations provide a rational basis for the design of future clinical trials employing morphometric end-points.

Immune-mediated neuropathy and myopathy in post-streptococcal disease: electron-microscopical, morphometrical and immunohistochemical studies

A 22-year-old man suffered from a complete flaccid tetraparesis and an immune complex-mediated rapid progressive glomerulonephritis after group A streptococcal infection. Serum creatine kinase was excessively elevated and myoglobinuria occurred. Nerve conduction studies revealed evidence of axonal neuropathy. Recovery was satisfactory within 18 months. Sural nerve and peroneus muscle biopsies were performed in the 4th and 14th week of the disease. Light microscopy of the sural nerve showed an incipient axonal type of neuropathy in the first biopsy. Ultrastructurally, Wallerian degeneration and endoneurial inflammatory cells were present. In the muscle biopsy, few atrophic fibers and altered blood vessels without further anomalies were found. In the second sural nerve biopsy, macrophages were numerous, some of which were immunoreactive for HLA-DR, and only a few myelinated and some unmyelinated nerve fibers remained. Muscle fibers in the second biopsy showed high-grade atrophy and myofibrillar abnormalities. Immunohistochemistry revealed diffuse endoneurial immunoglobulin deposition in the first sample, while in the later biopsy specimen, deposits of IgG, and kappa and lambda light chains were visible in circumscribed endoneurial areas. Immune-mediated neuropathy and myopathy are not well-known complications of streptococcal disease. This is, to our knowledge, the first detailed report on morphological findings in muscle and nerve in such a disorder.

Occlusive microangiopathy by immunoglobulin (IgM-kappa) precipitation: pathogenetic relevance in paraneoplastic cryoglobulinemic neuropathy

Histological, immunohistochemical and ultrastructural sural nerve and skin biopsy findings in a case of cryoglobulinemia secondary to an IgM-kappa-producing non-Hodgkin lymphoma are described. The main finding was an occlusive microangiopathy present in both the sural nerve and the skin. Widespread cryoglobulin deposits of the proliferated vasa nervorum were associated with pronounced changes probably evoked by ischemia. Moderate perivascular inflammation, but no florid vasculitis was additionally present. Our observations indicate that occlusive microangiopathy by precipitated cryoglobulins may be a relevant pathogenetic factor in cryoglobulinemic peripheral neuropathy.

Mitochondrial abnormalities in human sural nerves: fine structural evaluation of cases with mitochondrial myopathy, hereditary and non-hereditary neuropathies, and review of the literature

Fifteen cases of mitochondrial myopathy, three cases of hereditary motor and sensory neuropathy (HMSN) VI, and 280 cases of neuropathies of different etiologies were examined by electron microscopy for the presence of mitochondrial abnormalities in the sural nerve. Altered mitochondrial were found in most cases of mitochondrial myopathy, in all cases of HMSN VI, and in 25 cases out of the series of unselected neuropathies. The mitochondrial changes comprised enlargement with an amorphous matrix and distorted cristae, with hexagonal paracrystalline inclusions, and with prominent cristae containing oblique striations, and a variety of rare changes. Most mitochondrial abnormalities were found in Schwann cells. An increase of the number of mitochondria was noted in smooth muscle and endothelial cells of epineurial arterioles of two cases with mitochondrial encephalomyopathy. Neuropathy was present in all cases of mitochondrial myopathy according to morphometrical analysis. Whether neuropathy is caused directly by mitochondrial dysfunction or by other pathogenetic mechanisms remains to be determined.

Sensory nerve pathology in amyotrophic lateral sclerosis

A detailed morphometric study was performed on sural nerve biopsies to determine the consistency of sensory nerve pathology in amyotrophic lateral sclerosis (ALS) and to seek a correlation between the severity of peripheral nerve pathology and disease duration. Nerve biopsies from patients with ALS consistently showed evidence of early axonal atrophy, increased remyelination and a shift in the diameter distributions curve towards smaller fiber diameters. Importantly, the severity of sensory nerve pathology in ALS patients correlated with disease duration. The peripheral nerve sodium pump concentration of patients was not reduced. It is concluded that an ingravescent dorsal root ganglion neuronopathy is seen in the incipient stages of ALS, preferentially affecting the largest neurons and resulting in turn in progressive axonal atrophy, secondary demyelination-remyelination and finally in nerve fiber degeneration. Etiologically, a parallel involvement of motor and sensory neurons suggests a more widespread metabolic disturbance in ALS than simply "sick" motor neurons.

Endoneurial microvessels in human diabetic neuropathy. Endothelial cell dysjunction and lack of treatment effect by aldose reductase inhibitor

Endoneurial microvascular abnormalities have been invoked in the pathogenesis of diabetic distal symmetric polyneuropathy. Detailed morphometric analysis of the endoneurial microvasculature was correlated with previously published data on nerve fiber morphometry and teased fiber analysis obtained from the same sural nerve biopsies. Biopsy specimens from neuropathic diabetic patients were obtained before and after 12 mo of aldose reductase inhibitor (ARI) treatment and compared to 15 carefully age-matched control subjects. Diabetic microvessels showed basement membrane thickening and loss of endothelial cell tight junctions. Microvascular density and the frequency of microvessels closed by endothelial cells increased with age in diabetic and control nerves and were unaffected by diabetes. The density of microvessels showing patent lumina did not differ between control and diabetic subjects and was not related to age or diabetes. Closed microvessels were composed of postcapillary venules that were otherwise devoid of ultrastructural abnormalities. We suggest that microvascular closure by endothelial cells may be a physiological condition and is unlikely to have any pathogenetic significance in diabetic neuropathy. Based on the current limited biopsy material, we conclude that 12 mo of ARI treatment that induced significant fiber repair and regeneration had no detectable effect on endoneurial microvascular abnormalities. These data suggest that endoneurial vascular pathology is not a rate-limiting factor in fiber damage or repair at this stage of diabetic neuropathy.

Familial amyloidotic polyneuropathy type 1 in Kumamoto, Japan: a clinicopathologic, histochemical, immunohistochemical, and ultrastructural study

Seventeen autopsy and five biopsy cases of familial amyloidotic polyneuropathy were examined clinicopathologically, histochemically, immunohistochemically, and ultrastructurally. In the autopsy cases, amyloid deposits were predominant in the peripheral nerve tissues, autonomic nervous system, choroid plexus, cardiovascular system, and kidneys. Amyloid involvements in the anterior and posterior roots of the spinal cord, spinal ganglia, thyroid, and gastrointestinal tract were also frequent. In the cardiac conduction system, amyloid deposition was prominent in the sinoatrial node and in limbs of the intraventricular bundle. In the sural nerve biopsy, besides amyloid deposits, degenerative changes of nerve fibers and Schwann cells were detected ultrastructurally, and the morphometric analysis showed a marked reduction in the number of myelinated fibers which correlated with the clinical stage. Amyloid deposits were resistant to pretreatment with potassium permanganate in Congo red staining, and transthyretin was confirmed immunohistochemically as a major component of amyloid deposits, along with the presence of serum amyloid P-component. Besides the amyloid deposits, transthyretin was proven in the liver cells, epithelial cells of the choroid plexus, and pancreatic islet A cells, suggesting that the transthyretin produced by these cells is secreted, transferred into tissues, and deposited in situ as the major component of amyloid in this disorder.

[Advanced stage of giant axonal neuropathy]

A woman has appeared normal during her neonatal and childhood period except for a short stature. Her parents were healthy and non-consanguineous. At the age of 8, she noticed difficulty in climbing stairs and had tendency to fall. In her 13th year, she developed marked scoliosis and genu valgum. Physical examination at 14 years of age revealed a gentle and shy child of short stature with brown-black kinky hair. Neurological examinations revealed progressive mental retardation, optic nerve atrophy, moderate and coarse nystagmus on lateral and vertical gaze, atrophic tongue with fasciculations, slow and scanning speech, distal muscular weaknesses with diffuse atrophies in the four extremities and sensory deficiencies in all modalities with a glove-stocking type distribution. At the age of 15, she was unable to walk without a wheelchair. During the course she showed slowly progressive muscular weakness, ataxia and decreasing sensation especially in the lower extremities. She died of infection of the respiratory and urinary tracts at the age of 25. Pathologically the abnormalities in the biopsied and autopsied sural nerve were characterized by an advanced stage of nerve fiber degeneration without giant axons. The phrenic nerve obtained at autopsy at 1 to 10 cm from axon terminal revealed the presence of several large focal axonal swelling of 15-20 microns in diameter. On the other hand, sections of the phrenic nerve at 15 cm from axon terminal displayed a mild to moderate reduction in the number of myelinated fibers without giant axons. The difference of pathological findings among these specimens seems to depend on the time as well as the site of the examination.(ABSTRACT TRUNCATED AT 250 WORDS)

Experimental vitamin E deficiency in rats. Morphological and functional evidence of abnormal axonal transport secondary to free radical damage

Morphological and functional studies have been performed on experimental vitamin E deficient rats. The predominant morphological change was axonal dystrophy and degeneration in the rostral parts of the dorsal columns, particularly in the gracile fasciculi. The dystrophic changes comprised focal axonal swellings containing accumulations of normal and abnormal organelles which included tubulovesicular structures probably derived from the smooth endoplasmic reticulum, mitochondria, dense lamellar bodies, neurofilaments, multifascicular bodies and lysosomes. Similar but lesser changes were observed in distal peripheral nerves. The appearances suggested a disturbance of axonal transport with a defect of 'turnaround' in the distal axons. Studies on the axonal transport of endogenous acetylcholinesterase showed an impairment both of fast anterograde and retrograde transport. The changes were considered to be secondary to the lack of the antioxidant effect of vitamin E as the neurological deficits could be reduced by the concomitant dietary administration of the synthetic antioxidant ethoxyquin and were markedly aggravated by the administration of polyunsaturated fatty acids. It is suggested that the neurological syndrome produced by vitamin E deficiency could be the result of damage to the function of mitochondria and other intra-axonal membranous structures which would interfer both with fast anterograde transport and 'turnaround' and lead to a distal axonal degeneration.

Morphological changes in unmyelinated nerve fibres in the sural nerve with age

Quantitative changes in unmyelinated nerve fibres in sural nerves obtained at autopsy were evaluated in 28 normal adults. The following conclusions were reached. (1) The density of unmyelinated axons showed no significant correlation with age, but the densities of (2) Schwann cell subunits with axons, (3) Schwann cell subunits without axons, (4) single protrusions of Schwann cells and (5) collagen pockets, and (6) the mean number of Schwann cell profiles per axon, all showed positive correlations with age. Additionally, (7) the percentage of subunits containing unmyelinated axons and (8) the mean number of axons in single axon-containing Schwann cell subunits demonstrated negative correlations with age. The density of Schwann cell nuclei related to unmyelinated fibres did not show a significant change with age. The age-dependent changes in unmyelinated nerve fibres thus mainly consist of an increased production of processes by Schwann cells in the absence of cell multiplication. A decrease in unmyelinated nerve fibre density or a compensatory increase of small unmyelinated axons did not occur in these normal adults. In terms of relative sensitivity for the detection of the earliest changes in unmyelinated fibres, the indices (6) and (7) are considered to be useful and superior to the conventional assessment of unmyelinated axon density and diameter distribution. These two indices are not influenced by postmortem swelling of the axons and Schwann cells. Measurements of unmyelinated axon density and size distribution will continue to be useful in the assessment of more advanced pathological conditions.

[Polyneuropathy caused by trichlorfon: report of a case with electrophysiologic and histopathologic study of the sural nerve]

The authors observed a patient who worked in a farm and suffered an organophosphate intoxication (trichlorfon). The immediate effect was manifested by vomiting and abdominal cramps. Three months later he presented a distal symmetric sensorimotor (predominantly motor) neuropathy with distal muscle atrophy. Electromyography has revealed denervation changes in every muscle studied and the sensory and motor nerve conduction was slowed in arms and legs. The sural nerve biopsy specimen studied by light microscopy with semi-thin section and teased fiber preparation showed axonal degeneration. The ultrastructural studies of the axonal alterations consisted of degeneration of the neurofilaments and the neurotubules with granular appearance of the axoplasm.

Congenital demyelinating motor and sensory neuropathy with focally folded myelin sheaths

Six patients (5 index cases and 1 sib) with a congenital motor and sensory neuropathy are described. The clinical, genetic and electrophysiological features resembled Dejerine-Sottas disease or hereditary motor and sensory neuropathy (HMSN) type III. Sural nerve biopsy of 5 patients revealed segmental demyelination and remyelination with hypertrophic changes, although onion bulbs were not as ubiquitous as in classical HMSN type III. A striking discriminating feature from HMSN type III was an abundance of focal myelin thickenings (tomacula) present in nearly all teased fibres. Possible pathogenic implications are discussed. These cases corroborate the heterogeneity of congenital motor and sensory neuropathies.

A new perspective on myelinated nerve fibre pathology in experimental diabetes

Quantitative electron-microscopical techniques were used to investigate the effects of experimental streptozotocin-diabetes on the longitudinal morphology of myelinated fibres in rat sural nerve. The morphology of myelinated fibres was symmetrical at proximal and distal levels in the nerves of control animals aged three months. Further development of myelinated fibres in control animals between three and seven months of age involved proximal and distal fibre growth with the simultaneous formation of a proximo-distal taper. The same pattern of fibre growth was not observed in animals with experimental diabetes. Although both fibre and axon size increased with age in both proximal and distal regions of sural nerve in the diabetic animals, a taper was not established. This impairment of fibre maturation by experimental diabetes was manifest more strongly in the proximal part of the nerve. There was no evidence favouring selective involvement of the axon or Schwann cell in the fibre pathology found in the diabetic animals. The results of this study do not support previous suggestions that experimental diabetic neuropathy is mainly a dying-back axonopathy. This report reaffirms that normal maturation of myelinated fibres is prevented by experimental diabetes mellitus.

Quantitative histologic study of the sural nerve in Lesch-Nyhan syndrome

The sural nerve of a 31-year-old man with Lesch-Nyhan syndrome obtained at autopsy was studied histologically. Large, myelinated nerve fibers were reduced in number and no myelinated nerve fibers larger than 5 microns were seen. Diameter distribution of myelinated nerve fibers did not demonstrate a bimodal pattern. The density of myelinated nerve fibers was 5,530/mm2 and was decreased as compared to the controls. On electron microscopic examination, lipid-like inclusions were observed in the cytoplasm of some Schwann cells. The role of these inclusions could not be elucidated, but reduction of larger myelinated nerve fibers suggests a peripheral nervous disorder in patients with this syndrome; therefore, patients with Lesch-Nyhan syndrome must be reappraised for disorders of the peripheral nervous system.

Rett syndrome: findings suggesting axonopathy and mitochondrial abnormalities

We report the histopathologic findings of 3 sural nerve biopsies and 1 muscle biopsy from 3 patients with Rett syndrome. The 3 sural nerve biopsies demonstrated a few ultrastructural abnormalities, including the presence of many Pi-granules and mitochondrial changes in the cytoplasm of Schwann cells, occasional bands of Büngner and onion-bulb formations, and mitochondrial alterations in myelinated axons. Morphometric analysis disclosed reduction in the number of large myelinated fibers with normal densities in comparison to those of an age-matched normal control. Light microscopic examination of the biopsied muscle from a 6-year-old patient with Rett syndrome revealed the existence of many small, dark, angulated fibers with NADH-TR staining. Ultrastructural investigation of the muscle confirmed the presence of the dumbbell-shaped mitochondria. Peripheral nerve involvement and the possibility of mitochondrial abnormalities in Rett syndrome were suggested by the results.

Relationship of endoneurial capillary abnormalities to type and severity of diabetic polyneuropathy

Endoneurial capillary abnormalities have been assessed quantitatively in sural nerve biopsies from diabetic patients with different syndromes of sensory polyneuropathy: chronic painful neuropathy, newly presenting painful neuropathy, and painless neuropathy associated with neurotrophic foot ulceration. Comparisons were made with age-matched nondiabetic control subjects. The diabetic groups showed no abnormality in capillary density or mean endoneurial area per fascicle. Compared with control subjects, all diabetic patients had an increase in mean capillary diameter, capillary wall thickness, and outer tunic (basement membrane and pericytes) thickness. The increase in wall thickness was most pronounced in patients with painless neuropathy (200%) and less marked in similar patients with painful neuropathy (100%). The pericyte volume fraction of the outer tunic was reduced in all diabetic patients, implying that basement membrane hypertrophy and reduplication were responsible for outer tunic thickening. There was evidence of endothelial cell hyperplasia rather than hypertrophy. There was a correlation between the degree of basement membrane thickening and the severity of myelinated fiber abnormality assessed neurophysiologically and morphologically. This study shows a link between the degree of endoneurial capillary basement membrane thickening, the type of neuropathology, and the clinical expression of neuropathy in diabetes mellitus.

Familial multiple symmetric lipomatosis with peripheral neuropathy

We describe coexisting peripheral neuropathy and multiple symmetric lipomatosis in 4 of 7 siblings. The absence of either condition in 3 other generations of this family suggests autosomal recessive inheritance. None of the affected siblings were alcoholic, a factor some have proposed to explain the frequent occurrence of peripheral neuropathy in sporadic multiple symmetric lipomatosis. Serum lipid studies, including apoprotein A levels, were normal. Sural nerve biopsy from 1 patient showed nerve fiber loss, predominantly affecting large myelinated fibers. The relationship between myelin sheath thickness and axon diameter was normal, arguing that this neuropathy is not due to primary axonal atrophy.

Association of painful and painless diabetic polyneuropathy with different patterns of nerve fiber degeneration and regeneration

We evaluated neuropathological abnormalities in sural nerve biopsies from 6 nondiabetic control subjects and 16 age-matched diabetic patients with different syndromes of sensory polyneuropathy (6 with chronic painful neuropathy [CPN], 4 with newly presenting painful neuropathy [NPN], and 6 with painless neuropathy associated with recurrent neurotrophic foot ulcers [RFU]). Although all but one of the evaluated features of myelinated and unmyelinated fiber pathology could be found in every diabetic patient, certain myelinated fiber abnormalities were associated with the clinical characteristics of the neuropathy. Thus, myelinated fiber density was severely reduced, "empty" Schwann tubes (an index of myelinated fiber degeneration) were increased, and early regeneration (bands of Büngner [BB], nonmyelinated axons) was pronounced in the RFU group. Progression from BB to regenerating myelinated fiber cluster (myelination and maturation) was more successful in patients with CPN and NPN than in those with RFU, and the finding of fibers with disproportionately large Schwann cells (cytoplasm and myelin) relative to axon caliber was exclusive to patients with neuropathic pain. We concluded that 1) unequal rates of successful fiber regeneration may underlie the apparent difference in the extent of myelinated fiber loss between painful and painless diabetic polyneuropathy; 2) myelinated and unmyelinated fiber degeneration and regeneration per se are probably not the cause of neuropathic pain in diabetic polyneuropathy, because each occurred in patients with RFU; and 3) axonal atrophy may be involved in neuropathic pain generation.

Demyelinating changes in sural nerve biopsy of patients with HTLV-I-associated myelopathy

We describe the sural nerve pathology in 3 patients with HTLV-I-associated myelopathy (HAM). These patients showed remarkable symptoms of myelopathy and had high HTLV-I titers in the serum and CSF. The common pathologic findings in the sural nerves were as follows: slightly decreased density in myelinated fibers (6,567/mm2, 6,488/mm2, and 7,159/mm2; control 9,999 +/- 3,446/mm2), frequent occurrence of demyelinated and remyelinated fibers, and many degenerating fibers with globule-like myelin changes. Teased-fiber analysis indicated the globule-like changes (Dyck's G change) to be accompanied by the formation of adjacent demyelinated segments. The globules result from slowly repetitive degenerative changes of myelin and are considered to be a form of demyelination. However, they have seldom been found in other demyelinating neuropathies, indicating the demyelination process with globule formation in HAM to be rather specific for this disease.