Alcoholic neuropathy

Journal of the neurological Sciences, 1975, 25:333-346 ~ Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

333

Alcoholic Neuropathy An Electron-microscopic Study GIOVANNI TREDICI AND MARIO MINAZZI lnstituto di Anatomia Umana Normale, Universitci di Milano, and Divisione Neurologiea, Ospedale Generale, Vimercate ( Ita(v ) (Received 18 December, 1974)

INTRODUCTION

Peripheral neuropathy is a frequent complication of chronic alcoholism. Despite the wide occurrence of this form of neuropathy there are still differences in opinion about its pathogenesis and the nature of the pathological lesions involving the nerve. Thiamine deficiency has been considered the main pathogenetic factor (DennyBrown 1958; Victor 1958), but other nutritional deficiencies and even the direct toxic action of alcohol have been claimed to play a role in inducing the nerve lesions (Fenelly, Frank, Baker and Leevy 1964). Electrophysiological studies on groups of alcoholic patients with different clinical evidence of peripheral neuropathy (Mawdsley and Mayer 1965; Bergamini, Gandiglio, Fra, Bergamasco, Bram and Mombelli 1965 ; Wanamaker and Skillman 1966; Perfetti, Ferro-Milone and Pacchiani 1967; Blackstock, Rushworth and Gath 1972; Casey and Le Quesne 1972) have given equivocal results and have led the authors to hold contrasting opinions on the nature of the pathological process underlying the electrophysiological abnormalities. Mawdsley and Mayer (1965) considered segmental demyelination the most probable lesion accounting for the mild reduction in nerve conduction velocity that they observed, while Blackstock et al. (1972) were of the opinion that axonal degeneration was the main pathological event. Conflicting opinions have also been drawn from histopathological studies. Segmental demyelination has been reported by Gombault (1886), Greenfield and Carmichael (1935) and Denny-Brown (1958). Dyck, Gutrecht, Bastron, Karnes and Dale (1968) in a case studied with the electron microscope and Walsh and McLeod (1970) in 11 cases studied with the teased-fibre technique have shown evidence of axonal degeneration. In an electron-microscopic study of 11 cases of alcoholic neuropathy Bischoff (1971) reported either primary myelin sheath disease or axonal degeneration or even a mixed form. Lastly, Lascelles (quoted by Thomas 1971) found evidence of segmental demyelination secondary to axonal lesions.

lower and upper limbs paresthesiae, lower limbs anesthesia

M

M

43

49

53

Case 4

Case 5

Case 6

-

lower limbs hypaesthesla, paresthesiae

M

49

Case 3

lower limbs paresthesme, hypaesthesia

lower limbs anesthesia and algoparesthesiae

mild distal weakness

lower limbs distal paresis

mild distal weakness

mild distal weakness

distal weakness

lower limbs paralysis

motor

mild distal muscle hypotrophy

mild distal muscle hypotrophy

mild distal muscle hypotrophy

mild distal muscle hypotrophy

distal muscle hypotrophy, skin dystrophy

distal muscle hypotrophy, pellagroid skin

trophic

Neuropathy

normal

abolished knee and ankle jerks

abolished knee and ankle jerks

brisk knee jerk. reduced ankle jerk

abolished knee a n d ankle jerks

abolished knee and ankle jerks

reflexes

42.9

6.3

~,5.6

38.2

6.7

5.0

45.5

40:0

35.8

velocity ( m / s e c i

5.5

7.0

6.0

latency (m.s~

M o t o r nerve conduction ~

Motor conduction: medial popliteal nerve. Normal value: latency 6.2 ms ( S . D . l . l l : velocity 44.1 m,,sec (S.D. 3.5). b Sensory conduction: median nerve. Normal value: latency to peak 3.0 ms IS.D. 0.401: amplitude 8 35 IW,

M

i t~wer limbs paresthesiae

M

62

Case 2

upper limbs paresthesiae, feet, legs anesthesia

M

61

Case 1

sensory

Sex

Age (yr)

Case

C L I N I C A L A N D E L E C T R O P H Y S I O L O G I C A L F I N D I N G S IN A L C O H O L I C PATIENTS

TABLE 1

4.0

4.0

h4

5.0

3.5

................. amplitude{ltl/ i

3.9

latencj' ,'msJ

Sensory nerve ~onduction h

N

> ,<

2

~e

4a

A I , C O H ( ) I . I C N E U R ( ) I ' A I1 IY

335

These conflicting reports prompted the present morphological and morphometric electron microscope study, with the aim of ascertaining the nature of the degenerative and regenerative processes which involve the peripheral nerves in the course of chronic alcoholism. MATERIAL AND METHODS

Sural nerve biopsy (Dyck and Lofgren 1966) was obtained from 6 chronic alcoholic patients with different clinical and electrophysiological signs of peripheral neuropathy. Other obvious causes of polyneuropathy were excluded after appropriate investigations. The main clinical and electrophysiological data of the patients,are summarized in Table 1. All the patients had a long history of regular heavy alcoholic intake (more than 1½ 1 wine per day). Dietary intake was poor in all 6 cases but vitamin supply was apparently adequate, except in Case 1 whose daily diet was mainly constituted by maize. In all cases, a gradual evolution of neuropathy over a period of months was reported. Case 3 presented with mental confusion, and signs of peripheral nerve involvement were detected only on examination. No significant illness was reported in any case except in Case 5, who 7 years earlier was treated with streptomycin and PAS for pulmonary tuberculosis. However, no signs of nerve involvement were noted during the course of this treatment. Nerve biopsy was obtained 2-3 weeks after admission. Small nerve pieces were immediately fixed in 2 ~ glutaraldehyde in 0.12 M phosphate buffer for 2 hr, postfixed in OsO4 in phosphate buffer for 2 hr, dehydrated in ethanol and embedded in Epon 812. Semi-thin sections, 1-2 #m thick, were stained with Toluidine blue and viewed through the phase microscope in order to estimate the external diameter of myelinated fibres. In each case, more than 500 fibres from unselected areas of 3 or more fascicles were measured at an enlargement of x 1000. Three different unselected areas of each fascicle were studied by electron microscope. At least 3 fascicles were studied in each case. Histograms of unmyelinated fibres, the number of myelinated and unmyelinated fibres and number of nuclei per unit area were determined by cross-sectional micrographs (final enlargement, x 9000). Percentages of the various components of the nerve were determined with a quadratic lattice of lines. Control material was obtained from early autopsy of 2 men aged 16 and 52 yr respectively, who died as a result of traumatic brain injury. No previous significant illness had been reported in their medical files. RESULTS

In all 6 cases the typical aspect under phase microscope and in low power electron micrographs was that of chronic partial denervation with a reduction in the number of myelinated fibres and normal appearances in most of the surviving ones (Figs. l and 2). Only a few fibres showed signs of degeneration or regeneration. However, clusters of myelinated and unmyelinated fibres supported evidence of a previous rearrangement in nerve structure. ( l ) Myelinated fibres Myelinated fibres undergo degenerative and regenerative processes. Macrophagic

336

~;. I'REDI('I, M. MINAZZI

Fig. 1. Case 2. I ,,~-power electron micr,~gn~ph. No myelinated fibres are observed in this field: Ntimer~tl~ normal anmychn;~ted fibres are intermingled with denervated bands of the unmyelinated type (arrows). x 7500.

changes in Schwann cells with splitting and breaking down of the myelin which forms ovoids or clumps, and with almost complete dissolution of the axoplasm were the most frequently observed degenerative changes (Fig. 3). Less commonly, disorganization and breaking down of microtubuli and neurofilaments, an increased

ALCOHOLIC NEUROPATHY

337

Fig. 2. Case 1. Low-power electron micrograph showing decreased density ofmyelinated fibres. A myelinated fibre is cut through a large myelin loop. x 3750.

number and size of mitochondria and vesicles of the endoplasmic reticulum, clumps of lysosome-like particles and areas of dissolution of the axoplasm were seen. Concomitant with the axon involvement, myelin changes were apparent. Separation of inner layers of the myelin from the axolemma, splitting of the myelin lamellae and irregularities of the myelin sheath sometimes occurred in fibres with mild signs of axon involvement. Moreover, nodes of Ranvier occasionally showed detached loops and watery vesicles, which split the myelin lamellae. These changes were sometimes observed in apparently normal axons, but longitudinal sections generally showed patchy degenerative changes along the length of the axon. Fibres showing complete shrinkage of the axon were seen. The myelin might still appear well pre-

338

~;. 'FREDICI, M. M1NAZZI

Fig. 3. A d v a n c e d stage of W a l l e r i a n - h k c d c g e n e r a u n n of a large m \ e l i n a t e d fibre. A d e n e r v a l e d b a n d ,~i m y e l i n a t e d type i~ s h o w n (top leftL ~ 4500.

served, although collapsed, and therefore thicker than usual when compared to the minute cylinder remnant. Denervated bands (Fig. 3) made up of Schwann cell processes of irregular shape and devoid of axons, surrounded by the basement membrane, were sometimes observed. The organization, shape, saze and sometimes the content of myelin remnants generally allowed differentiation of this type of denervated bands from those derived from unmyelinated fibres. Apart from redundant fotdings m collapsed fibres (Fig 4), the basement membrane did not appear to be involved in pathological changes. Regeneration was evident by the appearance of small sprouts of axons devoid of myelin among the Schwann cell processes of the denervated bands. Few fibres of small size showed disproportionately thin myelin, loose lamellae with interposed cytoplasmic proces~s of the Schwann cells, indicative of an early stage of remvelination. The Schwann cells of these fibres may show an activated nucleus and electron-dense cytoplasmic matrix with an increased number of organelles. Clusters of 3-4 small myelinated fibres, generally 2-4 um in size and seldom larger than 6 #m, were often observed. Sometimes fibres devoid of myelin were collected into the clusters. Schwann cell processes and unmyelinated axons rarely form concentric layers around a single myelinated fibre or a small cluster resembling a small onion bulb formation. Very few fibres showed demyelinated segments or widening of the nodal gap.

(2) Unmyelinated fibres Unmyelinated fibres also showed degenerative and regenerative changes. Disorgani-

ALCOHOLIC NEUROPATHY

339

Fig. 4. Schwann cell devoid of axons. The hascmcnt membraneis redundant and shows numerousfoldings. x 26.000. zation of the tubular and filamentous structures of the axoplasm or even a complete breakdown of all its organized structures with a watery appearance of the axon were sometimes observed. Much more frequently bands of denervation (Pig. 1) were seen, made up of several layers of plate-like processes of the Sch~vann cells and completely devoid of axons. During regeneration, small round sprouts (0.2-0.4 #m) with an electron-dense matrix appeared among the Schwann cell bands. Often only the round shape permitted these regenerative growths to be distinguished from Schwann cell processes (Fig. 5). Degeneration and regenerative changes, both in myelinated and in unmyelinated fibres, occurred concomitantly. Thus in low power fields, one observed normal fibres together with fibres in various stages of degeneration and regeneration, and denervated bands adjacent to clusters of fibres which had completed regeneration.

340

(;. T R E I ) I ( ' I , M. M 1 N A Z Z I

Fig. 5. Denervated band of the unmyelinated type with several regenerative sprouts (arrows). ~: 22,501).

TABLF 2 QUANTITATIVE HISTOLOGICAl, MFASUREMENTS OF SURAL NERVE: VALUES PER U N I ] AR|-~

Nerve

.

N e r v e f~hres ( No. mtn'-j . . . . .

.

.

.

.

N u c l e i ~ Nt~. m m ~ : . .

myelinated

unmyelinated

Schwann

[~brohtast

other

Case 1

4740

39620

1970

330

130

Case 2

2t40

58590

3640

430

130

Case 3

950t)

49020

2730

5 H~

450

Case 4

4200

53050

3060

640

250

Case 5

7470

47850

2740

230

150

Case 6

7090

41160

2840

380

180

Control 1

9670

41130

1990

340

! 41)

Control 2

10120

49730

2780

420

210

Control a

10056

47009

2557

238

130

(7713 14373)

(32179-68813)

(1458-.-4167)

(81 4568)

Range a

aAverage values from 8 healthy nerves (Dyck, Lambert and Nichols 1972).

(0-534

341

ALCOHOLIC NEUROPATHY

(3) Other nerve compone, ts The autophagic changes in the Schwann cells during degeneration and their activation in the early stages of myelin deposition have already been reported. Similar changes were not observed during degeneration and regeneration in Schwann cells of the unmyelinated fibres. In other respects this group of cells become hypertrophied (see Quantitative data) and showed an increased tendency to encircle connective tissue forming collagen pockets. Normal morphology of collagen, interstitial tissue, fibroblasts, perineurium and vasa was observed in all cases. (4) Quantitative data In Table 2 the number of myelinated and unmyelinated fibres per mm 2 and the number of nuclei per mm 2 are reported, and in Table 3 the percentage ofmyelinated fibres of small and large size (more than 6/tm) cut through the nucleus is given. The number of nuclei per unit area was in the normal range. The number of myelinated fibres was reduced in all cases except Case 3, which appeared both clinically and histologically to be the least involved. Unmyelinated fibres were in the normal rang& The highest numbers of unmyelinated fibres were observed in Cases 2 and 4, the histograms of which show the highest incidence of small size fibres, indicative of regenerative activity. The percentage of myelinated fibres cut through the nucleus was not increased except in Case 2. TABLE 3 P E R C E N T A G E OF M Y E L I N A T E D FIBRES C U T T H R O U G H T H E N U C L E U S

Nerve

Small.fibres

Lar oe .fibres

Total

Case 1

5

1.5

6-5

Case 2

7

2

9

Case 3

4

2

6

Case 4

3

1.5

4.5

Vase 5

3

1

4

Case 6

4

1

5

Control 1

4

1

5

Control 2

4

2

6

(%)

(%)

o~ (,,)

The histograms of myelinated fibres (Fig 6) show bimodal distribution, but a shift to the left of the peak of small and large fibres is observed, indicating that the pathological process involved both types of fibre. In all cases except Case 2, an increase in the percentage of fibres up to 3/~m in size is evident. The histograms of unmyelinated axons (Fig. 7) show a bimodal distribution in Cases 1, 2, 4 and 6 in contrast with the unimodal distribution of healthy nerve. In all cases except Case 3, there is an increase in the percentage of fibres less than 0.6 #m in diameter.

342

(;. rRI{I)ICI,

3o1 25+ ~2ot

M.

MINAZZI

CONTROL.

16y

.~t

0 2 4 6 8 1012Hm CASE 1

~ o2 5, i

,~ ~

CASE 2 ~'-

!

2~o+ u~25 I"

A

~

105-

CASE 3

~5~

~ 25+

~2o ~ ~

~

10+ i ~ " ~ , / / ~ 5 ~

0 2 4 6 81012Hm

~

0 2 4 6 81012~Jm

CASE 4

f /

~

CASE 6

~01

0 2 4 6 81012Um

in

0 2 4 6 8 I012}Jm

CASE 5

~o

~-~

F/)~

10 } [,,/~/J~"~

g

2o~

0 2 4 6 81012pm

o"2',,~'6'~,'lb'l'2pm

Fig. 7. Histograms showing fibre size of unmyelinatedaxons. Fibres have been divided into 0.2film groups:

30t

CONTROL 16y

25}

t~ 15

~3 21 0 o ~ 6 5 O

CASE 1 25~m E ~5+VA

CASE

~30 i

5 2o~ 5

52y

pm

CASE 2

CASE 3

~

~ 25,

2O E 15

o ~5

~

~se 5 ~30 t

~ 2o~ 5

48y

m

CASE 6 ~3Of

53y

t~ 2Q

o. . . . . . . . .

Fig. 6. Histograms showing fibre size of myelinated axons. Fibres have been divided into 1-/~mgroups. In Table 4 the percentage of unit area of the endoneurium covered by each of the main c o m p o n e n t s of the nerve is given. In alcoholic patients there is a clear-cut increment in collagen and interstitial tissue and a proportional reduction in the area covered by myetinated and unmyelinated axons. The myelinated axon c o m p o n e n t and the myelin component are reduced in all cases: the value of the Schwann cells in

343

ALCOHOLIC NEUROPATHY

TABLE 4 P E R ( ' E N T A G E OF E N D O N E U R I U M COVERED IN CROSS-SECTIONS BY DIFFERENT NERVE COMP()NENTS

Nerve

Colla~ten and interstitial tissue (o~)

Myelinated a wms (%)

Myelin " (<',i)

Mvelinated S c hwann cells (,!~>)

Unmyelinated axons (o~)

Unrnyelinated

Schwanncells (?,,)

Case 1 (40540)"

65

6

13

3

2

10

Case 2 (40480)"

70

2

6

1

1.5

17

Case 3 (31720)"

55

9

19

4.5

3.5

9

Case 4 (35820)"

62

3

10

3

2.5

15

Case 5 (41420)"

53

8

19

4.5

2.5

13

Case 6 (39320)"

61

5

t4

4.5

2.5

13

Control 1 (50640)"

44

13

26

3.5

3.5

9

Control 2 (42330)"

47

12

25

4.5

3.5

8

~Area examined expressed in itm 2. Fibroblasts and vasa are included under the heading Collagen and interstitial tissue.

myelinated fibres is generally unchanged, but is decreased in Case 2, in which the reduction of the myelinated axon component is particularly important. The axon component of the myelinated fibres appears to be more involved than the myelin component as indicated by the ratio of the percentage myelin-percentage myelinated axon, which is 2 in normal cases and is increased in nerves of alcoholic subjects, despite the reduction in the myelin component. The unmyelinated axon component is reduced in all cases except Case 3, while conversely the Schwann cell component of the unmyelinated fibres is increased in all except Case 3.

DISCUSSION

In all 6 alcoholic patients reported in this study, the nerve shows the typical aspects of chronic partial denervation due to Wallerian-like degeneration of the fibres. No evidence to support the possibility of primary myelin sheath disease (Greenfield and Carmichael 1935; Bischoff 1971) has been obtained. The observed changes in the myelin sheath appear to be secondary to the axon lesion, as evidenced in longitudinal sections. Furthermore, they are similar to the reported changes of myelin in the early stages of experimentally-induced Wallerian degeneration (Nathaniel and Pease 1963; Ballin and Thomas 1969 ; Williams and Hall 1971). Similar changes have also been

344

{;, ['REDICI, M. MINAZZI

shown to occur frequently in 2 cases of uraemic neuropathy (Dyck, Johnson. Lambert and O'Brien 197l) in which clear-cut signs of secondary segmental demyelination have been demonstrated. Since we have nol undertaken single teased-fibre studies, we cannot give a definite statement on the occurrence and amount of secondary segmental demyelination in our cases. However. the number of demyelinated segments or wider nodal gaps did not exceed that reported in normal cases (Lascelles and Thomas 1966 : Ochoa and Mair 1969b), in accordance with the data of Walsh and McLeod I 1970}who. in a teased-fibre study of I1 alcoholic patients, did not report increased segmental demyelination. Different myelin involvement probably might occur in single cases of alcoholic neuropathy in relation to the individual course of the illness or to other unidentified factors (e.g., the type of drink and the drinking and dietary habits}. Further evidence of primary axonal involvement can be seen in the changes in the unmyelinated fibres, which are in all respects similar to those observed in experimentally-induced WaIlerian degeneration of unmyelinated nerves {Dyck and Hopkins 1972). Moreover, the analysis of the quantitative data gives further support to our opinion about primary involvement of the axon. In fact. the ratio of myelin component to myelinated axons is increased in pathological nerves despite the reduction in myelin, indicating a greater involvement of the axonal component. In many forms of neuropathy caused by deficiency or metabolic disorders, histological studies have often presented conflicting results (Thomas. Hollinrake. Lascelles. O'Sullivan, Bailtod, Moorhead and Mackenzie 197t}. Such difficulties in assessing the nature of the nerve lesion are mainly due to the chronic course of the disease and the patchy and asynchronous involvement of the fibres. Moreover. the early changes in the myelin, similar in both axonal degenerauon and in primary demyelination (Allt and Cavanagh 1969; Ballin and Thomas 1969 : Dyck et al. 1971 }, indicate the necessity to judge the type of pathological involvement from fibres in a fairly advanced stage of degeneration, and the importance of electron-microscopic study (with the aid of longitudinal sections) for detecting early axonal lesions. The concomitant involvement of axons and Schwann cells even in early stages of degeneration is not surprising if one considers the biological connections linking the axon and the related Schwann cells. Besides the well-known importance of axon size in controlling deposition and thickness of myelin {Friede 1973), there is probably a mutual metabolic dependence between neuronal cells and Schwann cells {Dyck et al. 1971; Bradley and Williams 1973). Mild derangement in axonal flow. evidenced pathologically by alterations in filamentous and tubular structures of the axoplasm. may therefore cause an inadequate metabolic supply to the Schwann ceils, evidenced by changes in the myelin sheath. Regeneration was evident in the wide occurrence of myelinated and unmyelinated fibres of unusually small diameter (Ochoa and Mair 1969b) and by the presence of clusters of fibres. The regenerative processes observed in our patients are m many respects similar to those observed in isoniazid neuropathy in man (Ochoa I970) and in unmyelinated fibres to those reported in the rat in sympathetic trunks (Dyck and Hopkins 1972). Degenerative and regenerative changes occur simultaneously in different fibres. The nerve appears therefore to be involved in a continuous process of degeneration and regeneration of its fibres, not unlike that which occurs much less

ALCOHOLIC NEUROPATHY

345

extensively under normal conditions (Lascelles and Thomas 1966; Ochoa and Mair 1969 a, b). Nutritional deficiencies and toxic factors seem to increase the destructive phase and reduce the regenerative capacity of the nerve, leading clinically to the gradual appearance of motor and sensory defects and pathologically to the aspects of chronic denervation. The grade of pathological involvement of the nerve in the course of chronic alcoholism was determined by the quantitative estimation of the different nerve components. The increase in connective tissue and reduction of the axon component was evident, compared to the value of either Control 1 (16 yr old) or Control 2 (52 yr old). Although many of the pathological changes reported in our study may be partly due to ageing (Ochoa and Mair 1969b), the extent of denervation in alcoholic patients greatly exceeds that previously reported in normal ageing and that determined in our older control. One must therefore assume the occurrence of active degenerative processes. Despite the reduction in the axon component, the number of myelinated and unmyelinated fibres was not always greatly reduced. This fact is explained by the increased number of small-sized fibres as shown by histograms of myelinated and unmyelinated fibres, indicating active regenerative processes. The percentage of myelinated fibres cut through the nucleus was in the normal range. This value has been considered an index of the length of the internodes (Ochoa and Mair 1969 a, b) and it is expected to increase in re~nyelinating fibres. The normal values observed may be due to the relatively few fibres actually involved in pathological changes. The morphometric data evidence loss of balance between degenerative and regenerative processes that the nerve undergoes even under normal conditions. This derangement, induced by metabolic and nutritional deficiencies, seems to be the pathogenetic basis of the neuropathy observed in the course of chronic alcoholism.

SUMMARY

The sural nerves of 6 patients with different signs of alcoholic neuropathy were studied qualitatively and quantitatively by electron microscopy. Myelinated and unmyelinated fibres showed degenerative changes of the Wallerian type. Concomitant involvement of the myelin appears to be secondary to axonal lesions. Regenerative processes, although frequently observed, did not balance the destruction of fibres in the degenerative phase. Quantitative studies indicated a reduced number of myelinated fibres and decreased percentage of the area covered in cross-section by myelinated and unmyelinated axons. The histograms of myelinated fibres showed a shift to the left of the peak of large and small fibres and an increased number of small-sized fibres. Similarly the histograms of unmyelinated fibres showed a shift to the left and a bimodal distribution with an increased number of small-sized fibres. Imbalance in degenerative and regenerative processes seems to be the basis of the chronic partial denervation observed in the nerves of alcoholic patients in this study.

346

(;. FREI)IC1, M. MINAZZI

REFERENCES ALLT. G. AND J. B. CAVANAGH 119691 Ullrastructurat changes in the region of the node of Ranvier in the rat caused by diphtheria toxin. Brain. 92: 459--468. BALL1N. R. H. M AND P. K. THOMAS I1969) Changes a~ nodes of Ranvier during Walterian degeneration An electron microscope study, A c t a neuropath. (Berl.), 14 237-249. BERGAM1N1. L.. G. GAND1GL10. L. FRA, B. BERGAMASCO. S. BRAM AND A. M. MOMBELL1 11965 }Alterazioni della conduzione nervosa sensitiva e m o t o n a in alcoolisti cronici privi di segni clinici di neuropatia periferica. Riv. Pat. nerr. ment., 86:31 49. BISCHOFF. A. (1971} Die alkoholische Polyneuropathie. Klinische ultrastrukturelle und pathogenetlsche Aspekte, D t s c h rned. Wschr.. 9 6 : 3 1 7 322. BLACKSTOCK. E.. G. R~SHWORTH AND D. GA'rH 119721 Electrophysiological studies in a l c o h o l i s m . . ! Neurol. Neurosur¢t. Psychiat.. 35: 326-334. BRADLEY, W. G. AND M. H. WILLIAMS(1973) Axoplasmic flow in cats with toxic neuropathles. Brain, 9b: 235 246. CASEY, E. B, AND P. M. LE QUESNE (19721 Electrophysiological evidence for a distal lesion m alcoholic neuropathy, J. Neurol. Neurosurg, Psychiat,. 35: 624-630. DENNY-BROWN. D. (1958)The neurological aspects of thiamine deficiency, bed. Proc.. t 7 (Suppl. 21:35 39. Dvcrz. P. J. AND A. P. HOPKINS ( 1972} Electron microscopic observations on degeneration and regeneration of unmyelinated fibres. Brain. 9 5 : 2 2 3 234 DYC'K. P. J. AND E. P. LOFGREN [ 19661 Method of fascicular biopsy of huma n peripheral nerve 1"orelectrophysiologic and histologic study, Proc. M a y o Clin.. 41 : 778--784. DYer, P. J., E. H. LAMBERT AND P. C. N~C~OLS [1972} Quantitative measurement of sensation related to c o m p o u n d action potential and number and sizes of myetinated and unmyelinated fibers of sural nerve in health, Friedreich's ataxia, hereditary sensory neuropath:r ans tabes dorsalis. In: A. RI~MOND (Ed.), Handbook o f Electroencephaloqraphv and Clinical Neurophysiolofl v, Vol. 9. Elsevier Publishing Company, Amsterdam pp. 83 -118. DYCK. P. J., W. J. JOHN~)N. E. H. LAMBERTAND P. C. O'BRIEN 11971 ) Segmental demyelination secondary to axonal degeneration in uraemic neuropathy, Proc. M a y o Clin._ 46: 4010431. DYCK, P. J.. J. A GUTRECH"I-. J. A. BASTRON. W. E. KARN~ AND A. J D. DALE (1968 Histologic and teased-fibre measurement of sural nerve in disorders of lower motor and primary sensory neurons. Proc M a y o Clin.. 43 : 81-123. FENELLY, J., O. FRANK. H BAKER AND C. M. LEERY 11964) Peripheral neuropathy of the alcoholic. Part I (Aetiological role of Aneurin and other B-complex vitaminsl, Brit, reed..I.. 2: 1290- t292. FRIFa~E, R. L. [19731 Control of myelin formation by axon caliber Iwith a model of the control mechanism I, ,I. comp. Neurol., 144: 233-252. GOMBAtJLT. M. [ 1~861 Sur les I&ions de la n+vrite alcoolique. C. R. 4cad. S c i ( D ) Paris J 102 : 439-440 GREENEII~LD. J. G. ~ND E. A. CARMICHAEI 11935} The peripheral nerves in cases of subacute combined degeneratmn of the cord, Brain, 58: 483-489, LASCELLES, R. G. ANl) P. K. THOMAS 119661 Changes due to age in internodal length in the sura] nerve in man, J. Neurol. Neurosurq. Psvchiat,. 29: 40..44. MAWDSLEY. C. ANrJ R. F. MAYER 11965] Nerve conduction in alcoholic polyneuropathv, Brain. 88: 335 356. NATHANIEL. E. J. H. AND D. C. PEASE [ 1963 I Degenerative changes in the rat dorsal roots during Wallerian degeneration, J. UItrastruet. Res._ 9 : 511-532 OCHOA, J. (19701 Isoniazid neuropathy in man Quantitative electron m~croscope study, Brain. 93: 831-850. OCHOA, J. AND W. G. P MAIR [1969a) l'he normal sural nerve in man. Parl 1 [Ultrastructure and numbers of fibres and cells l, A eta neuropath. Berl. ). 13 : 197 -216. OCHOA. J. AND W. G. P. MAre f1969bt The normal sural nerve in man. Part 2 [Changes in the axons and Schwann cells due to ageing I, ~Iota neuropath (Berl.), 13:217 239. PERFETTL C . C . . F. FERRO-MtLONE AND A. PACCmAN! (t967) Studio dei valori di_dispersione della velocit~ di conduzione [VDC} deUe fibre motorie negli alcoolisti. Riv. Neurol., 37: 453-461. THOMAS, P. K. 11971 ] The morphological basis for alterations in nerve conduction in peripheral neuropathy, Proc. roy. Soe Med. 64: 295-298. THOMAS, P. K.. K. HOLLINRAKE. R. G. LASCELLES. D. J. O'SULLIVAN. R. A. BA1LLOD. J. F, MOORHEAD ANO L C MACKENZIE [1971 ) The polyneuropathy of chronic renal failure, Brain, 94: 76L 780. VICTOR. M. [ 1958) Alcohol and nutritional diseases of the nervous system, J. Amer. reed. Ass.. 167: 65-71. WALSH. J. C AND J. G. McLEoD 11970 ~, Alcoholic neuropathy An electrophysiologicat a nd histological study, J. m,urol. Sci.. 10:457 469, WANAMAKER. W. M. AND T. G. SKILLMAN {19661 Mo t or nerve conduction in alcoholics. Quart. J. Stud, Alcohol. 27: 16-22. WILLIAMS. P. L, AND S. M. HALL (1971] Chronic Wallerian degeneration An in vivo and nltrastructural ~tudy. J. Anat ( L o n d . 109:~R7 503