ELSEVIER Journal of the Neurological Sciences 127 (1994) 68-76
JOURNAL OF THE
NEUROLOGICAL SCIENCES
Epidemic neuropathy in Cuba: morphological characterization of peripheral nerve lesions in sural nerve biopsies
Israel Borrajero a , l Juan Luis P6rez a, Carlos Domlnguez a, Agustln Chong a, Rosa Maria Coro a, H6ctor Rodriguez b, Nelson Gomez c, Gustavo C. Romfin d,,,
Lydia Navarro-Romfin e
a National Reference Center for Pathologic Anatomy, Laboratory of Pathology, Havana, Cuba b Departments of Internal Medicine, Havana, Cuba
c Neurology, Hermanos Ameijeiras Hospital, San Lazaro 701, Centro Habana, Havana, Cuba, d Neuroepidemiology Branch, National Institute of Neurological Disorders and Stroke, Bethesda, MD, USA
e Laboratory of Pathology, Hematopathology Section, National Cancer Institute, National Institutes of Health (NIH), Bethesda, MD, USA
Received 4 March 1994; revised 6 July 1994; accepted 8 July 1994
Abstract
More than 50000 patients were affected in Cuba during an epidemic outbreak of peripheral neuropathy from January 1-992 until September 1993. The disease presented as either a retrobulbar optic neuropathy, a predominantly sensory peripheral neuropathy, a dorsolateral myeloneuropathy, or as mixed forms. The morphological findings in sural nerve biopsies from 34 patients with various forms of the disease are presented here. Frozen, paraffin and semi-thin sections were prepared for light and electron microscopy, immunohistochemistry and morphometric analysis. Every case presented morphological alterations ranging from mild axonal dystrophy (9 cases, or 27%) to moderate and severe axonal damage (25 cases, or 73%): In 6 cases (18%), axonal damage was accompanied by perineural fibrosis and vascular abnormalities. Axonal regeneration was noted in 8 cases (23%) and remyelination in 9 (26%). Morphometric analysis showed a predominant loss of myelinated fibers in 92% of the patients. Quantification of myelinated fiber loss in 11 patients revealed a remarkable decrease in large caliber fibers. Scarce mononuclear cells were observed in 17 cases. No virus-like elements were seen. The morphological features found in this study indicate that, regardless of the clinical presentation, peripheral nerve lesions of the epidemic neuropathy in Cuba correspond to an axonal neuropathy. These lesions are compatible with nutritional, toxic, or metabolic etiologies. An inflammatory etiology would be unusual with these lesions.
Tropical myeloneuropathies (TMN), optic and pe- ripheral neuropathies have been reported periodically in the Caribbean during the last century (reviewed by Rom~in et al. 1985, 1986). Nonetheless, the magnitude of the outbreak of epidemic neuropathy in Cuba is exceptional. From January 1, 1992, through January 14, 1994, the Cuban Ministry of Public Health identified
1 TO whom requests for reprints should be addressed. * Corresponding author. At: 343 W. Houston St., Suite 204, San
Antonio, TX 78205, USA. Tel.: (210) 226-8349; Fax: (210) 227-3918.
and treated 50,862 cases of neuropathy (MMWR 1994). This subacute neurological syndrome was character- ized by retrobulbar optic neuropathy alone or accom- panied by a predominantly sensory distal and symmet- rical neuropathy, by dorsolateral myeloneuropathy, sensorineural hearing loss, dysphagia or dysphonia (Roman 1994). Four main clinical syndromes were ob- served: (1) Retrobulbar optic neuropathy manifested by decreased visual acuity, loss of color vision, central or cecocentral scotomata, pallor of the temporal aspect of the optic disks and loss of axons in the maculopapil- lary bundles. (2) Peripheral sensory neuropathy mani- fested by burning pains and dysesthesias in fingers, soles and toes, with excessive sweating, pins-and-nee-
I. Borrajero et al. / J o u r n a l o f the Neurological Sciences 127 (1994) 6 8 - 7 6 (~t)
dles sensations, and decreased or absent ankle reflexes. (3) Dorsolateral myelopathy presenting with signs of pyramidal tract involvement such as brisk reflexes in upper and lower limbs with crossed adductor responses but without Babinski sign, as well as with proximal leg weakness and symptoms suggestive of spastic bladder, such as urinary frequency and nocturia, along with signs of dorsal columns involvement, including sensory ataxia in severe cases. (4) Mixed forms with several combinations of the above, as well as signs of cranial nerve involvement, mainly sensorineural deafness, dys- phonia and dysphagia. In most cases, neurological symptoms were preceded by weight loss and fatigue.
Multiple clinical and epidemiological research stud- ies were launched to clarify the etiology and pathogen- esis of the disease. As part of these investigations, sural nerve biopsies in well-documented cases were per-
formed at the "Hermanos Ameijeiras" Hospital, the National Reference Center for Pathology in Cuba. In the past, only rare morphological descriptions of the peripheral nerve lesions observed in TMN have been published (Williams and Osuntokun 1969; Labrousse 1986). We present here the results of sural nerve biopsies studied during the outbreak in Cuba. Wc characterized the morphological lesions as an axonal neuropathy.
2. Material and methods
Case selection Thirty-four patients with a diagnosis of epidemic
neuropathy were selected, based on the diagnostic cri- teria defined by the Cuban Ministry. of Public Health
T a b l e 1
S u m m a r y o f c l in ica l a n d p a t h o l o g i c a l f i n d i n g s in p a t i e n t s w i t h e p i d e m i c n e u r o p a t h y in C u b a
C a s e Age Sex E v o l u t i o n O . N . S e n s o r y D o r s o l a t . D e a f n e s s D y s p h a g i a P a t h o l o g i c a l R e g e n e - M a c r o - M F D
N o , (y r s ) n e u r o p a t h y m y e l o - d y s p h o n i a c l a s s i f i c a t i o n r a t i o n p h a g e s
n e u r o p a t h y
1 50 M 8 m o n t h X X X X A N * - + 213~
2 26 F 9 m o n t h X X X A N - ~ + + 152{)
3 74 F 24 m o n t h - X A N - + + 175S
4 62 M 1 m o n t h X X A N + + + 2260
5 68 M 2 m o n t h X X A N * + + 1561
6 37 F 16 m o n t h X X A N - - 2175
7 57 F 4 m o n t h X A N - 2913
8 45 F 5 m o n t h X X A N - 31)61
9 48 M 1 m o n t h X X A N - + IS~ ~)
11) 55 M 6 m o n t h X X X A N - + (~76
11 40 F 4 m o n t h X X A N + - IN(~3
12 67 M 4 m o n t h X X A N * - - 21197
13 34 F 1 m o n t h - X A D - -- 311~
14 61 M 2 m o n t h X X A N - + 131t~
15 44 F 2 m o n t h - X A N - +
16 20 F 8 d a y s - X X A D - • 3821
I7 42 F 6 m o n t h X X X X ,AN + + - 2686 18 27 F 4 m o n t h X X X A N + + + 11)24
1 t) 40 F 1.5 m o n t h X X X A D - -- 4343
20 33 M 1 m o n t h - X A D - -
21 52 M 4 m o n t h X X A N + + 1179
22 34 F 14 m o n t h X X A N + ~ 52t)
23 45 F 1 m o n t h X X X X X A D + + -- 421)8
24 73 F 16 m o n t h X X A N * - - 132
25 63 F 9 m o n t h X X X X A N * - - 3 [ (~0
26 33 M 2 m o n t h X X X A N * - - 5073
27 49 M 7 m o n t h X X A D - -
28 26 F 3 m o n t h - X A D - - S 150
2~ 41 F 15 d a y s X X X A D - - 5455
30 39 F 3 m o n t h X X X X A N - +
31 47 F 8 m o n t h s X X A N - +
32 27 F 1 m o n t h s X X A N - -
33 44 F 3 m o n t h s X X A N - -- +
34 26 F 15 d a y s - X A D - S~'~ 75
* E n d o n e u r a l f i b r o s i s a n d v a s c u l a r c h a n g e s .
O . N . = O p t i c N e u r o p a t h y
M F D = M y e l i n a t e d F i b e r D e n s i t y / m m ~
A N = A x o n a l N e u r o p a t h y
A D = A x o n a l D y s t r o p h y
71) I. Borrajero et al. / Journal ~f' the Neurological Sciences 127 (1994) 68- 76
and by the Pan American Health Organizat ion/World Health Organization mission to Cuba ( P A H O / W H O 1993). Patients were admitted to the "Hermanos Amei- jeiras" Hospital for study and treatment.
256 gray levels and a TV camera (B/W). Fascicular area and myelinated fiber density were measured in 26 patients. In addition, myelinated fiber diameters were measured in 11 patients.
Sural nerce biopsies Biopsies were performed by the method of Dyck et
al. (1984) after informed consent was obtained from all patients. Nerve biopsy fragments were either snap- frozen in OCT (Gurr, BDH Laboratory Supplies, Eng- land) for immunohistochemical studies, or fixed in 10% buffered formalin for paraffin embedding. Glutaralde- hyde fixation followed by osmic acid postfixation and araldite embedding were used for semi-thin and elec- tron microscopy preparations.
Histological, electron microscopy, and immunohis to- chemical studies
Paraffin sections. Longitudinal and transverse sections were obtained from formalin-fixed paraffin-embedded nerve tissue stained with hematoxylin-eosin ( H & E ) and Masson's trichromel In addition, Luxol-fast blue (LFB) stained sections and neurofilament (DAKO, Glostrup, Denmark) preparations were studied. Sec- tions were also stained with LCA (CD 45, DAKO A / S , Glostrup, Denmark). Immunoperoxidase technique was performed following the ABC method (Hsu and Raine 1981).
Frozen sections. Frozen sections (5-8 /~m) were pre- pared, acetone-fixed and stained with the panleukocyte marker LCA (CD 45), and for OKIal, HLA class II (Ortho, Raritan, N J). Immunoperoxidase staining (Hsu and Raine 1981) was completed using an avidin-bio- t in-horseradish peroxidase technique (ABC complex, DAKO, Glostrup, Denmark), using rabbit anti-mouse immunoglobulin and 3,3'-diaminobenzidine free base (Sigma) as the developing solution, and hematoxylin as counterstain.
Thin and semi-thin sections. Araldite-embedded sec- tions were prepared according to a previously pub- lished technique (Benington 1978). Transverse semi- thin sections were stained with p-phenylenediamine (Lampert and Scochet 1979). Thin sections were con- trasted with lead citrate and uranyl acetate and studied under a Hitachi H-7000 electron microscope.
Morphometric studies. Morphometric analysis was per- formed on semi-thin sections prepared as above. A DIGIPAT System for image analysis (EICISOFT- CUBA 1991, Centro de Software y Robotica, Cuba) was used for this purpose. This system uses a CPU 386 33 MHz (640 k b / 4 Mb memory), digitizing card B / W
3. Results
The median age was 44 years with a range from 20 to 74 years. Five patients were over 60 years old. Eleven patients (32%) were male and 23 female (68%) with a male / female ratio of 1:2. Table 1 summarizes the demographic data and clinical findings in these patients. Seven cases (20%) presented with isolated peripheral sensory neuropathy, 5 (15%) presented a dorsolateral myeloneuropathy and the remaining 22 (65%) had combined optic and peripheral neuropathy. 15 patients were referred from Havana, 12 from Pinar del Rio, 5 from Santiago de Cuba. 1 from Villa Clara. and 1 from Cienfuegos.
Histological and immunohis tochemical findings"
Paraffin sections. Changes observed in light microscopy are summarized in Table 2. The most frequent findings were mixoid endoneurial degeneration, endoneurial fi- brosis, and perineural fibrosis. The latter change was found in cases with a mean evolution of 9 months and it occurred mainly in patients older than 60 years of age. Immunohistochemical staining with LCA showed mononuclear cells scattered in the neural interstitium and perivascular spaces, without vasculitis in 17 of 28 (61%) cases. OKIa (HLA class II) was positive in Schwann ceils in 19 of 25 (76%) cases. Staining for neurofilaments disclosed irregularities or loss in the continuity of axons, as well as axonal thickening, in 17 of 22 analyzed cases.
Semi-thin and thin sections. Results are summarized in Tables 3 and 4. All nerve samples showed morphologi- cal alterations consistent with an axonal neuropathy (AN), ranging in severity from mild damage or axonal dystrophy observed in 9 cases (27%), to modera te and severe lesions noted in 25 cases (73%).
Table 2 Light microscopy findings in 26 patients with epidemic neuropathy in Cuba
1. Borrajero et al. /Journal of the Neurological Sciences 127 (1994) 68-70 71
Fig. 1. Transverse section of sural nerve showing decrease of large and small diameter myelinated fibers. Notice scarcity of large-caliber myelinated fibers. Semi-thin section stained with p-phenylenediamine ( :x 200).
Axonal lesions were characterized by axonal loss (Figs. 1 and 4) with presence of Bungner bands (Fig. 2) along with signs of degeneration (Fig. 3) manifested by aggregates of degenerated mitochondria, endoplasmic reticulum dilation resulting in microcyst formation, dense bodies, loss of neurotubules, myelin bodies, and axonal sequestration. In 8 cases (23%) signs of axonal regeneration were seen. Remyelination was observed
in 9 cases (26%). Increased endoneurial fibrosis and more pronounced vascular alterations, mainly arterio- lar wall thickening with basal membrane reduplication, were observed in 6 patients (18%), These changes were most commonly seen in patients over 60 years of age.
Axonal dystrophy lesions were characterized by a lesser degree of the changes described above. In this group of patients, 6 of 9 cases presented with brief
4 \
Fig. 2. Sural nerve demyelination. A large-diameter fiber is demyelinated with formation of Bungner bands. Electron micrograph ( × 250(I).
72 I. Borrajero et aL /Journal of the Neurological Sciences 127 (1994) 68-76
Fig. 3. Schwann cell engulfing myelinated fiber with axonal degeneration, collapse, and disruption of myelin sheath. Electron micrograph ( x 3000).
duration of the disease (1 month or less) suggesting that these mild changes represent early stages of the axonal degenerative process. Of these 9 cases, 3 had a clinical picture of dorsolateral myeloneuropathy.
It must be emphasized that, in some cases, the process of axonal degenerat ion was accompanied by normal appearing myelin sheaths. In cases where the fibers were undergoing disintegration both the axon and the myelin sheath were encircled by a Schwann
cell (Fig. 3), or were phagocytized by endoneurial macrophages. Axonal degenerat ion with ovoid forma- tion was observed in both semi-thin and thin sections (Figs. 4 and 5). Ultrastructural search for viral particles or virus-like elements proved completely negative.
M o r p h o m e t r y . Fascicular area in the biopsies ranged from 36756 to 269139 /xm 2 with a mean of 109951 /xm 2. Normal values were found in 20 patients (be-
Fig. 4. Severe axonal degeneration with ovoid formation. Semi thin section stained with p-phenylenediamine ( × 200).
1. Borrajero et al. /Journal of the Neurological Sciences 127 (1994) 68 76 73
Fig. 5. Axoplasm degeneration and disruption of myelin with ovoid formation. Electron micrograph ( × 2500).
tween 60 000 and 190 000/ . tm 2) (Jacobs and Love 1985). Fou r pa t i en t s had va lues be low normal and two had h i ghe r - t han -no rma l values.
Mye l ina t ed f i be r density. ( M F D ) v a l u e s r anged from 132 to 8675 f i b e r s / m m 2, with a mean of 2894 f i b e r s / m m 2. However , the 9 cases with axonal dys t rophy had M F D
5
09
Z W
<
3
0 0
0 <i <2 <3 <4 <5 <6 <7 <8 <9
M Y E L I N A T E D F IBERS X l , 0 0 0 MM 2 N - 26
Fig. 6. Myelinated fiber density (MFD) in sural nerve biopsies in patients with epidemic neuropathy in Cuba. Note the shift towards MFD values below 3000/mm e observed in 16/26 cases (61%).
74 1. Borrajero et al. /Journal o( the Neurological Sciences 127 (1994) 68-70
Table 3 Light microscopy findings in semi-thin sections in 34 patients with epidemic neuropathy in Cuba
Morphologic change No. of patients
Signs of axonal regeneration 8 23 Presence of macrophages 11 32 Presence of ovoids 13 38 Lymphoid cells 3 9 Minor vascular alterations 21 62 Hyaline arteriosclerosis 1 3
values greater than 3000 with a median of 5000. In 16 cases out of 26 (61%) there was a noticeable decrease in MFD with values lower than 3000 fibers/mm 2 (Figs. 1 and 6). Eleven of the 14 cases with moderate to severe axonal neuropathy lesions presented MFD be- low 2000 (median 1000). Only 2 of the 26 patients studied showed MFD within normal limits (between 7500 and 10000 myelinated fibers/mm 2 of endoneurial surface) (Jacobs and Love 1985; Labrousse 1986) (Fig. 6). The latter 2 patients were young and had a brief disease evolution showing only changes of axonal dys- trophy.
Table 4 Electron microscopy findings in 34 patients with epidemic neuropa- thy in Cuba
Morphologic change No. of % patients
Aggregates of degenerated mitochondria 25 73 Intra-axonal laminated bodies 16 47 Dilation of endoplasmic reticulum 10 29 Intra-axonal dense bodies 22 65 Loss of neurotubules 18 53 Neurofilament dissolution 5 15 Complete degeneration of the axon and the sheath 17 50 Axonal sequestration 11 32 Signs of remyelination 9 26 Bungner's bands 22 65 Mucoid degeneration of the endoneural stroma 10 29 Increased collagen fibers t4 41
mately 40%) (Labrousse 1986), Much less significant was the decreased percentage of small-caliber fibers (lower than 6 ~m); values ranged between 28% and 74% with a mean of 53% (normal values around 60%) (Labrousse 1986). Figure 7 shows a typical fiber diame- ter histogram illustrating that myelinated fibers smaller than 9 tzm represent 49% of the total.
Myelinated fiber diameters. These were measured in 11 biopsies. Decreased percentage of large-caliber fibers (from 9 to 12 /xm) was documented ranging from 3% to 26% with a mean of 14% (normal values approxi-
4. Discussion
Despite the relatively high frequency of neu- ropathies and myeloneuropathies in the tropics, with
36
u~
IJJ =3 2o
U,. =It:
0 0 .......
~1 <2 <3 <4 <5 <6 <7 <8 '¢9 <10 <11 <12 :<13 <14
D I A M E T E R ( ) = r n )
N - 200 F I B E R S
1
Fig. 7. Myelinated fiber diameter distribution in a patient with epidemic neuropathy in Cuba. Notice decrease of large-diameter fibers (9-12/~m) with nearly normal distribution of small-diameter ( < 9 p.m) myelinated fibers.
L Borr¢(jero et aL /Jo , rnal o.~" the .,%k'z~roh~gictd Sc'i{'pt~'e~ 127 (1994) 6~¢- 70 75
prevalence figures higher than those of multiple sclero- sis in temperate regions, there is a dearth of neu- ropathological descriptions of these disorders. Mortal- ity from these conditions is relatively low and post- mortem studies are few. Nerve biopsies, which provide more information with fewer artifacts, are performed infrequently in tropical myeloneuropathies . The changes observed in this group of patients with epi- demic neuropathy in Cuba correspond to an axonal neuropathy. Although various degrees of myelin degen- eration were observed, the presence of fibers with normal myelin sheaths but with obvious axonal alter- ations are consistent with primary, axonal damage (Be- nington 1978; Bishop 1979). Likewise, some Schwann cells with marked phagocytic activity of degenerated myelin were observed, probably as an expression of the severe degenerative process undergone by the myeli- natcd fibers (Bcnington 1978).
Of the nine cases with minimal axonal fiber damage classified as axonal dystrophy according to Lampert and Scochet criteria (1979), three presented with dor- solateral myeloneuropathy. The above changes would suggest that the underlying lesion in the group of patients with dorsolateral myeloneuropathy may be a central-distal peripheral axonopathy (Spencer and Schaumburg 1976).
Light microscopy examination revealed frequent edema-mixed endoneurial degeneration, endoneurial fibrosis and thickening of the wall of small endoneurial arterioles. These changes were confirmed by electron microscopy and may be related to the axonal damage and secondary demyelination, lmmunohistochemical studies of neurofilaments (Chalk and Dyck 1992) were also consistent with the axonal alterations found by ultrastructure. Schwann cells were positive to HLA-an- tigen (class 11) suggesting that the degenerative process may have stimulated these cells to express the antigen (Mancardi et al. 1988). The scarcity of mononuclear cells observed and their perivascular and interstitial distribution were interpreted as a reactive process re- lated to axonal and myelin degeneration or to the normal immunological surveillance process, rather than to a primary damage (Feasby 1992; Hann et al. 1988; Koski 1992).
The decrease of MFD values reaching figures as low as 132/mm 2 and 5 2 9 / m m 2, is well below the lower limit reported by Labrousse (933 /mm 2) in cases of tropical peripheral neuropathies. Low MFD values were observed in 92c7c of the Cuban series suggesting that this measurement is important for the histological characterization of this neuropathy. In addition, there was a relationship between the severity of the axonal damage and the number of myelinated fibers: in the 9 cases with mild axonal dystrophy, the median MFD was around 5000 while in the moderate and severe cases the median MFD was about 1000. This decrease
is a direct expression of myelinated fiber loss. The marked reduction of large-caliber fibers is noteworthy, since these fibers are predominantly affected in most axonal neuropathies.
The axonal neuropathy we report here is consistent with the different stages of evolution of the peripheral nerve lesions caused by nutritional, toxic, or metabolic agents (Williams and Osuntokun 1969: Shiraki lt~75; Benington 1978). A viral origin would be unlikely bc- cause of the absence of viral elements by electron microscopy and the scant inflammatory infiltrates ob- served. Since the morphological alterations described are non specific it is impossible to establish the causal nature of the epidemic in Cuba solely on this basis.
The wide clinical spectrum of neurological syn- dromes observed during the outbreak in Cuba resem- bles closely the clinical manifestations described by Spillane (1947) after World War II. Fisher (1955) re- viewed the postmortem findings in thcse cases and found a selective loss of the maculopapillary tracts in cases with optic neuropathy. Lesions of dorsal columns probably due to a neuronopathy affecting dorsal root ganglia were also observed. Previous reports of nerve biopsies in the tropics also provide some information. Williams and Osuntokun (1969) published the first report of nerve biopsies in 23 patients with tropical ataxic neuropathy (TAN) from Nigeria. Thcsc authors concluded that the primary nerve lesion was segmental demyelination. Inflammatory lesions were often ob- served, as well as pcrineural collagen infiltrati~m. On electron microscopy a predominant loss of myclinated fibers was noted with some evidence of axonal changes. No signs of regeneration were observed. The most likely etiology of TAN in Nigeria is probably a combi- nation of malnutrition and excessive cyanide intake from exclusive consumption of cassava. Labrousse (1986) reviewed 29 biopsies in patients with tropical myeloneuropathies of unknown origin from Ivory Coast. In 95% of the patients an axonal ncuropathy was documented with selective loss of large-caliber myeli- hated fibers, quite similar to the lesions described here.
Among the nutritional conditions presenting with similar morphological changes, one of the most com- mon is beriberi, due to thiamin (vitamin B~) deficiency. In the past beriberi reached epidemic proportions throughout the tropics, but was noted to resurface recently in Japan (Takahashi 1976). Nerve biopsies in patients with beriberi (Ohnishi et al. 1~,~8(I) revealed an axonal neuropathy with formation of myelin ovoids, selective loss of large-diameter myclinated fibers, and formation of Bungner bands identical to the lesions described here. In most studies of TMN a nutritional or toxic origin, a combination of these two factors or an unknown etiology, is postulated (Lcmercier ct al. 1971; Corvisier c ta l . 1987).
7~ I. Borrajero et al. / Journal of the Neurological Sciences 127 (1994) 68-76
Although the exact cause of the epidemic neuropa- thy in Cuba remains elusive, sural nerve biopsies in patients with several clinical forms of the disease al- lowed characterization of the basic lesion as an axonal neuropathy, circumscribing the possible etiological agents to nutritional, toxic, or metabolic causes. The uniformity of the morphological lesions in patients with different clinical presentations would suggest that all the clinical forms observed in this epidemic may be part of a common pathogenetic mechanism.
Acknowledgements
The authors gratefully recognize the support of the Epidemic Neuropathy Task Force (Grupo Operativo Nacional, Cuba) and the Pan American Health Organi- zation/World Health Organization (PAHO/WHO: Dr. Miguel A. Marquez, local representative, Havana, Cuba). Mrs. Liliana Jimenez Vazquez provided secre- tarial support. The Cardiology Institute, Havana, al- lowed full time participation of Dr. R.M. Coro in this research. Dr. P.S. Spencer and Prof. P.K. Thomas reviewed the original biopsy materials and provided helpful advice. We are grateful to the technical staff of the Department of Pathology of the "Hermanos Amei- jeiras" Hospital, to the referring physicians, and to the patients for active participation in this study.
References
Benington, J.L. (1978) Pathology of Peripheral Nerve, W.S. Saunders Company, Philadelphia, PA, pp. 46-47, 50-95, 272-278.
Bishop, A. (1979) The peripheral nerve. In: Johansen, J.V. (Ed.), Electron Microscopy in Human Medicine, Vol. 6, Nervous sys- tem, sensory organs, and respiratory tract, McGraw-Hill Book Co., New York, NY, pp. 137-236.
Chalk, C.H. and Dyck, P.J. (1992) Application of immunohistological techniques to sural nerve biopsies. Neurol. Clin. Periph. Neu- ropathy, 10: 601-612.
Corvisier, N., J.M. Vallat, J. Hugon, M. Lubeao and M. Dumas (1987) Les neuropathies de cause ind&ermin~e. Rev. Neurol. (Paris), 143: 279-283.
Dyck, P.J., J. Karnes, A. Lais, E.P. Lofgren, and J.C. Stevens (1984) Pathologic alterations of the peripheral nervous system of hu- mans. In: Dyck P.J., Thomas P.K., Lambert, E.H. and Bunge R. (Eds.), Peripheral Neuropathy, W.S. Saunders Company, Philadelphia, PA.
Fisher, M. (1955) Residual neuropathological changes in Canadians held prisoners of war by the Japanese. (Strachan's disease). Can. Serv. Med. J., 11: 157-199.
Hahn, A.F., T.E. Feasby, A. Steele. D.S. Lovegren and J. Berry (1988) Demyelination and axonal degeneration in Lewys rat ex- perimental allergic neuritis depends on the myelin dosage. Lab. Invest., 59: 115-125.
Hsu, S.M. and L. Raine (1981) Protein A. avidin and biotin in immunohistochemistry. J. Histochem Cytochem., 29: 1349-1353.
Jacobs, J.M. and S. Love (1985) Qualitative and quantitative mor- phology of human sural nerve at different ages. Brain. 108: 897-905.
Labrousse, F. (1986) La biopsie nerveuse au cours de neuromy61opa- thies tropicales. A p r o p o s de 29 observations. Th~se pour le diplome d'~tat de docteur en m~decine. Universit~ de Limoges. Limoges, France.
Lampert, P.W. and S.S. Scochet (1979) Ultrastructural changes of peripheral nerve. In: Trump, B.F. and Jones, R.T. (Eds.), Diag- nostic Electron Microscopy, Vol. 2. John Wiley & Sons, New York, NY.
Lemercier, G., L. Jacquin-Cotton and H. Collomb (1971) Myelopathy of "obscure" etiology-nutritional syndrome. Neurology, 21: 185- 188.
Mancardi, G.L., A. Cadoni. A. Zicca. Q. Scheonne. M Tabaton. I. De-Martini and D. Zacheo (1988) HLA-DR Schwann celt reactiv- ity in peripheral neuropathies of different origins. Neurology, 38: 848-851.
Ohnishi, A.S., H. Tsujis, Y. Igisu, I. Muray, Y. Goto. M. Kuroiwa. M Tsuihata and M. Takamori (1980) Beriberi neuropathy. Morpho- metric study of sural nerve. J Neurol. Sci.. 45: 177-190.
PAHO/WHO Mission to Cuba (1993) Epidemic Neuropathy in Cuba. Epidemiol. Bull.. 14: 1-4.
Rom~in, G.C. (1994) An epidemic in Cuba of optic neuropathy, sensorineural deafness, peripheral sensory neuropathy and dorso- lateral myeloneuropathy. J. Neurol. Sci., 127: 11-28.
Rom~in, G.C., P.S. Spencer and B.S. Schoenberg (1985) Tropical myeloneuropathies: the hidden endemics. Neurology, 35: t158- 1170.
Shiraki, I. (1975) The neuropathology of subacute myelo-optico-neu- ropathy, "SMON", in humans. Jap. J. Med. Sci. Biol., 28 (Suppl.): 101-164.
Spencer, P.S. and H.H. Schaumburg (1976) Central-peripheral distal axonophaty: the pathology of dying-back polyneuropathies. Prog. Neuropathol., 3: 253-295.
Spillane, J.D. (1947) Nutritional Disorders of the Nervous System, Livingstone, Edinburgh.
Williams, A.O. and B.O. Osuntokun (1969) Peripheral neuropathy in tropical (nutritional) ataxia in Nigeria: light and electron micro- scopic study. Arch. Neurol., 21: 475-492.
