electrophysiological studies alcoholism - bmjtion in the peripheral sensory nerve fibres in these...

Journal of Neurology, Neurosurgery, and Psychiatry, 1972, 35, 326-334

Electrophysiological studies in alcoholismEILEEN BLACKSTOCK, GEOFFREY RUSHWORTH, AND DENNIS GATH

From the Unit of Clinical Neurophysiology, Department of Neurology, Churchill Hospital, Oxford,and the University Department ofPsychiatry, Warneford Hospital, Oxford

SUMMARY Using a range of electrophysiological techniques, it has been possible to demonstrateimpaired function in smaller calibre motor fibres and in distal large cutaneous sensory nerve fibres inboth alcoholic patients without neuropathy and in those alcoholics with clinical manifestations ofperipheral nerve disease. Evidence of more proximal involvement of Ia sensory fibres was obtained,but in the majority of our patients, large motor fibres functioned normally. The nature of the under-lying pathological process is discussed.

The study of the disorders of peripheral nervefunction has been greatly facilitated by thedevelopment and application of refined electro-physiological techniques. Recent advances inthis field, such as the introduction by Hopf(1962, 1963) of a method by which velocity inthe slower conducting, and hence, fine motornerve fibres can be measured, together with thedevelopment by Buchthal and Rosenfalck (1966)of a detailed technique for measuring cutaneoussensory nerve potentials and their maximumvelocities, have widened the scope of electro-physiological procedures.Some workers (Mawdsley and Mayer, 1965;

Juul-Jensen and Mayer, 1966), using the nowwell-established methods of measuring maxi-mum conduction velocities in peripheral motorand sensory nerve fibres in alcoholic patientswith polyneuropathy, have found small de-creases in these velocities. However, Perfetti,Milone, and Pacchiani (1967), using Hopf'smethod of double stimulation to examine theulnar nerve in alcoholics with polyneuropathyor other neurological disturbances, found thatthe large motor fibres conducted normally,whereas slowing of velocity was observed in thesmall fibres.

Similarly equivocal findings have been reportedfollowing electrophysiological investigation ofalcoholic patients who at that stage had noclinical evidence of peripheral nerve disturbance.Several workers (Jurko, Currier, and Foshee,1964; Mawdsley and Mayer, 1965; Bergamini,Gandiglio, Fra, Bergamasco, Bram, and Mom-belli, 1965; Wanamaker and Skillman, 1966)have reported marginal reductions in maximum

326

conduction velocities in peripheral motor nervesin such patients, whereas others (Juul-Jensenand Mayer, 1966; Vacek, Stika, Grofova, andWarman, 1969) have found maximum conduc-tion velocities in these nerve fibres to be withinnormal limits. Lack of agreement also existsbetween those who have studied nerve conduc-tion in the peripheral sensory nerve fibres inthese patients. Mawdsley and Mayer (1965)indicated that electrophysiological abnormali-ties occurred equally in peripheral sensory andmotor nerve fibres, whereas Bergamini et al.(1965), studying a small group of alcoholics,concluded that significantly greater electro-physiological abnormalities were detectable inperipheral sensory fibres.

It seemed worthwhile to investigate, using arange of electrophysiological techniques, bothalcoholics with peripheral nerve lesions andthose with no manifestations of alcoholic poly-neuropathy, in order to determine whether ornot electrophysiological abnormalities of motorand sensory nerve fibres were detectable. Wewere particularly interested in the measurementof cutaneous sensory nerve potentials and theirmaximum velocities in these patients, as we feltthat useful information about distal sensory nervefunction may be obtained by the application ofthis technique.

PATIENTS

The patients studied had been admitted to localpsychiatric hospitals (the Warneford Hospital, asmall acute psychiatric hospital, and the Ley Clinic,a unit specializing in the treatment of drug depend-ency) for treatment of alcoholism or its attendant

by copyright. on July 24, 2021 by guest. P

rotectedhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.35.3.326 on 1 June 1972. D

ownloaded from

http://jnnp.bmj.com/

Electrophysiological studies in alcoholism

TABLE 1CLINICAL DETAILS

Patient Physical findinigs Dietary Mental state onNo. Age Neglect admission

(yr) Muscle power and Sensibility Reflexes Othercoordination findings

1 35 Marked proximal and Defective appreciation, Absent Hepatic + + + Disorientation for time,distal muscle wasting light touch, pin prick, knee enlargement place, and person.with loss of power in vibration, and position and Emotionally labile. Im-arms and legs. Inco- sense in arms and legs ankle paired memory for recentordination in arms and jerks events. Short-termlegs memoty impaired

2 63 Moderate proximal and Defective appreciation Absent Hepatic + + + Disorientation for time,distal muscle wasting in light touch, pin prick, knee enlargement place, and person. Grosslegs. Power and co- and vibration sense and memory impairment forordination normal below the knee ankle recent and remote events

jerks

3 50 Normal Defective appreciation Normal Hepatic + + + Normalpin prick and vibration enlargementsense below knee

4 55 Normal Defective appreciation Absent Hepatic + + Normallight touch and pin prick ankle enlargementdistally in arms and legs. jerksAbsent vibration senselower limbs

5 55 Slight wasting proximal Defective appreciation of Absent Nil + Disorientation for time,and distal muscles in light touch and pin ankle place, and person. Visuallegs. Power and co- prick distally in arms jerk hallucinationsordination normal and legs

6 50 Normal Normal Absent Hepatic + + + Impaired memory forankle enlargement recent events. Impairedjerk short term memory

+ During month preceding admission one cooked meal daily. Snacks only at most other mealtimes.+ + During month preceding admission no cooked meals on most days.

+ + + During month preceding admission no cooked meals on most days with loss of weight exceeding 14 lb.

complications. Selection of patients took place in sofar as only patients who had at least one index ofsevere alcoholism-that is, early morning tremulous-ness relieved by further drinking, an abnormalpattern and frequency of drinking, alcohol amnesias,loss of control, and hallucinatory experiences-wereadmitted to the study. A single patient only showedjust one of these features. Also, patients who hadconcurrent diseases like diabetes, or other possiblecauses of peripheral nerve disease, were excluded.

After admission, a history, including details ofrecent drinking and dietary habits, was taken fromeach patient, using a semistructured interviewschedule. This was supplemented where possible byinformation obtained from relatives. A completepsychiatric and physical assessment was carried outon each patient.

Altogether 30 patients (23 males and seven females)were studied. The mean age of the group was 44-8years (age range 21 to 65 years). The average age ofcommencing regular drinking was 20 years (range13 to 37 years). More than half the patients (17)drank spirits (whisky, brandy, gin) exclusively; ofthe remainder, five drank sherry or cheap wine, two

preferred beer or cider, and six patients were inter-mittent or regular methyl alcohol drinkers. Eightpatients gave a history of severe dietary neglect withweight loss greater than 14 lb (6-4 kg) beforehospitalization. On physical examination eightpatients had hepatic enlargement; signs of peripheralnerve involvement, details of which are given inTable 1, were present in six patients (group I); theremaining 24 patients (group II) had no symptomsor signs of polyneuropathy.As part of another study, measurement of thia-

mine deficiency (using the pyruvate tolerance testand the red cell transketolase estimation) was madein 13 patients, three of whom had neuropathy.These tests were carried out as soon as possible afterhospitalization and before vitamin therapy wascommenced. Abnormal pyruvate tolerance testswere present in four patients, one of whom hadneuropathy, and in one of these four the red celltransketolase was also abnormal.

METHODS

All the tests were carried out in a warm room with

327by copyright.

on July 24, 2021 by guest. Protected

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.35.3.326 on 1 June 1972. Dow

nloaded from


Eileen Blackstock, Geoffrey Rushworth, and Dennis Gath

the patient lying on a couch and covered withblankets. All patients were asked to warm up theirarms by immersion in hot water for 10 minutesbefore the session. In all cases measurements weremade on the non-dominant side. The limbs on whichthe tests were carried out were supported on hotwater bottles throughout the procedure. The skintemperature of the distal part of the limbs wasmeasured at the end of the recording session, themean temperature being 33°C.

1. MEASUREMENT OF MOTOR NERVE MAXIMUM CON-DUCTION VELOCITIES Using coaxial needle elec-trodes, muscle action potentials were elicited fromthe abductor pollicis brevis after maximum electricalstimulation of the median nerve at the wrist, elbow,and in the axilla, and from the extensor digitorumbrevis by stimulating the anterior tibial nerve at theankle and the lateral popliteal nerve at the knee.The surface stimulating electrodes were of the Disatype, bipolar and approximately 5 mm in diameter,with an interelectrode distance of 25 mm. Beforestimulation at each site the skin was prepared bygentle sandpapering and application of electrodejelly. Square wave stimuli, derived from a DisaMultistim, were applied and repeated at the rate ofone every 2 sec. The muscle action potentials werefed into a high input impedance cathode follower,amplified by conventional means and displayed on adouble beam oscilloscope, one beam ofwhich was fedwith a time scale derived from a crystal oscillator(Digitimer, Devices Instruments Ltd). The muscleaction potentials were simultaneously photographedas sweeps (speed 2 mm/msec). Latencies weremeasured from the stimulus artefact to the beginningof the initial deflection from the baseline of thecompound potential. Distances between the variouspoints of stimulation and the recording sites weremeasured along the course of the nerves and dividedby appropriate latencies to give velocities.

2. MEASUREMENT OF CUTANEOUS SENSORY NERVEPOTENTIALS The digital nerves were electricallystimulated using two silver rings 4 mm wide, coveredby lint moistened with saline, and applied firmlyaround each finger in turn, with the cathode proxi-mal to the first interphalangeal joint and the anodeover the terminal interphalangeal joint. The surfacerecording electrodes, which consisted of silver probescontained in suction cups, with an interelectrodedistance of 2 cm, were placed exactly over the courseat the wrist of the median and ulnar nerves in turn,these points having been accurately mapped outwhen recording the muscle action potentials. Squarewave stimuli were applied and repeated at the rate of2/sec. The stimulus strength was adjusted to besupramaximal for the response. The sensory nerveaction potentials were amplified and displayed on adouble beam oscilloscope, as described above. The

sensory potentials were simultaneously photographedas 10 superimposed sweeps (speed 5 mm/msec) oncontinuously moving 70 mm photographic paper.The latency of the nerve action potential wasmeasured from the stimulus artefact to the beginningof the first positive deflection. Sensory nervevelocities were calculated by dividing the distance,measured on the skin, between the stimulatingcathode and the recording site, by the appropriatelatency.

3. MEASUREMENT OF H-REFLEX AND ANKLE JERK TheH-reflex was recorded from the soleus and gastro-

,., i.....

18

W..-

,iY4 fii

I

I

*-.4

FIG. 1. Measurement of minimum motor conductionvelocity using two supramaximal stimuli separated bysuccessively greater time intervals.

328

Al-Ilk

.1

i

.A

i".-.., -

.W

- f --



j.com/

J Neurol N


ownloaded from



6 7 8 9 10 11 12 13 14 15

TIME (m secs)

_ ________i msc

lTITlP, C

p v-_"

miw....I

FIG. 3. Cutaneous sensory nerve potential (digit 1) of a healthy control subject compared withthat of an alcoholic patient without neuropathy.

cnemius muscles by means of coaxial needle elec-trodes, after a submaximal stimulus had beenapplied to the medial popliteal nerve in the poplitealfossa. For the measurement of the latency of theankle jerk, a patella hammer containing a piezo-electric crystal was used. This, when struck, produceda potential change capable of triggering the cathoderay oscilloscope. The reflex electrical activity of the

calf muscles was recorded through needle electrodesas before.

4. MEASUREMENT OF MINIMUM MOTOR CONDUCTIONVELOCITIES IN ULNAR NERVE This procedure wascarried out on eight patients from group II. Hopf'smethod of double stimulation was used. Muscleaction potentials were recorded from the abductor

25

E

EE

CL1=

20

15

10

5

FIG. 2. Determination of thelate/icy of the slow motorfibres.

I%ft Im

329

i

Os.- :2



j.com/

J Neurol N


ownloaded from



digiti minimi using surface electrodes, one over themuscle belly and the other over the tendinousinsertion. Stimuli were delivered through two sets ofbipolar stimulating electrodes which were strappedfirmly to the arm, one over the course of the ulnarnerve at the wrist and the other above the elbow.The electrodes were also held throughout theprocedure to prevent them slipping. Both stimuliwere adjusted to give supramaximal responsesseparately. The separation in time was found atwhich the response to the second stimulus was justvisible. This was recorded by photographing five

50

40 -

minimum motor conduction velocity in the ulnarnerve was carried out on eight control subjects whowere matched as closely as possible for age withgroup II alcoholics on whom this measurement wasmade.

RESULTS

MAXIMUM MOTOR CONDUCTION VELOCITIES As itis recognized that maximum conduction veloci-ties in peripheral nerves may be slower in peopleover the age of 50 years (Norris, Shock, and

0@

0

I I I I

A

0

FIG. 4. Distribution of the amplitudeA of cutaneous sensory nerve potentials in

alcoholic patients without neuropathyand controls. A= control subjectmatchedfor age and sex. 0 = alcoholicwithout neuropathy.

I I I I I

0 2 4 6 8 10 12 14 16 18 20 22

AMPLITUDE OF SENSORY NERVE POTENTIAL OF FIRST DIGIT (Measured in microvolts)

superimposed sweeps. The time interval between thestimuli was then successively increased by 0-5 msecand the responses recorded in a like manner. Theseparation was continued up to 20 msec separation,to enable accurate determination of the timeinterval for the second response to become maximal.A graph was then constructed plotting the amplitudeof the successive responses to the second stimulusagainst the successive time intervals betweenstimulus I and II. Using the graph, it was possible todetermine the exact time interval at which thesecond response became maximal; this representedthe latency of the slow motor fibres. The distancebetween the two stimulating cathodes was measuredon the skin and the velocity calculated in the usualway.

5. CONTROL GROUP This consisted of 14 peopledrawn from the medical and nursing staff. Measure-ments of cutaneous sensory nerve potentials andtheir maximum conduction velocities, and thelatencies of the H-reflex and the ankle jerk weremade on each control subject. Measurement of the

Wagman, 1953), group II, those alcoholicpatients without polyneuropathy, was subdividedinto two age ranges: 21 to 50 years (17 patients),and 51 to 65 years (seven patients). Group I,

which consisted of six alcoholics with clinicalevidence of polyneuropathy (mean age 51l3years), was considered too small to be thus sub-divided.The mean values for the maximum conduc-

tion velocities of the median and lateral poplitealnerves for group I patients are presented inTable 2.

These mean values are within the limits whichare regarded as being normal in this laboratoryfor the segments of each nerve examined. Twopatients in this group deserve closer attentionhowever. Case 1, aged just 35 years, showedconsistently low maximum conduction velocitiesin the upper limb for someone of this age(axilla-elbow segment 47 m/sec, elbow-wristsegment 46 m/sec). This patient also had the

= 30

z

o 20

10

0

330

IA



j.com/

J Neurol N


ownloaded from



TABLE 2MAXMUM MOTOR CONDUCTION VELOCITIES IN

ALCOHOLIC PATIENTS WITH POLYNEUROPATHY (GROUP I)

Nerve Mean (± one SD) Range

MedianAxilla-elbow 61-3 ± 10-3 47-80Elbow-wrist 51-7±39 46-57Terminal delay 3-75 ± 05 3 5-4-5

Lateral poplitealKnee-ankle 47 ± 7*9 30-54Terminal delay 68 ± 14 5-5-9 5

Maximum motor conduction velocities expressed in m/sec.Terminal delay measured in msec.

most severe motor nerve involvement on clinicalexamination. Case 5, who on clinical examina-tion had signs of motor nerve disturbance in thelower limbs only, gave the lowest recordedmaximum conduction velocity in the knee-ankle segment (30 m/sec), together with anabnormally long terminal delay of 9-5 msec inthe anterior tibial nerve. Both patients hadstriking abnormalities of sensory nerve function;cutaneous sensory nerve potentials were un-recordable in both cases and the latency of theH-reflex was much prolonged (34 msec and40-5 msec).The mean values for the maximum conduction

velocities for patients without neuropathy(group II) are given in Tables 3 and 4.These mean values for both subgroups of

alcoholics without neuropathy are within accept-able normal limits, although certain patients areexceptional. In the subgroup under 50 years ofage, two patients showed reduced maximumconduction velocities in the knee-ankle segment

TABLE 3MAXIMUM MOTOR CONDUCTION VELOCITIES INALCOHOLIC PATIENTS WITHOUT NEUROPATHY

(GROUP II < 50 YEARS)


MedianAxilla-elbow 62-9 ± 7-8 48-75Elbow-wrist 56 1±4-5 48-62Terminal delay 3-8 ± 07 2-75-5-3

Lateral poplitealKnee-ankle 48-1 ± 7-2 33-58Terminal delay 6-9 ± 1-7 4 25-11


TABLE 4MAXIMUM MOTOR CONDUCTION VELOCITIES INALCOHOLIC PATIENTS WITHOUT NEUROPATHY

(GROuP H> 50 YEARS)


MedianAxilla-elbow 52-3±9-1 31-60Elbow-wrist 53-3 ± 58 43-61Terminal delay 3-7±04 3-14-3

Lateral poplitealKnee-ankle 48 ± 93 40-65Terminal delay 5-8± 1-5 3.5-7.5


of the lateral popliteal nerve, and prolongedterminal delay in the anterior tibial branch ofthis nerve (39 m/sec, 7-9 msec; 33 m/sec,11 msec respectively).

Similarly in the 51 to 65 years subgroup, onepatient had abnormally low maximum conduc-tion velocities in both segments of the mediannerve (axilla-elbow segment 31 m/sec; elbow-wrist 43 m/sec). These three patients also hadnotable electrophysiological abnormalities onexamination- of peripheral sensory nerves; theamplitude of the recorded sensory nerve poten-tials was reduced for all three and the latenciesof the H-reflex and ankle jerk abnormally long.

MINIMUM MOTOR CONDUCTION VELOCITIES Dis-played in Table 5 are the values for the maximumand minimum conduction velocities measured inthe ulnar nerve of eight alcoholic patientsdrawn from group II, compared with thosevalues determined for a group of eight controlsubjects. Two values for the minimum velocityare given in each instance. The first has beencalculated as previously described; for thesecond value, 1 msec, to allow for the refractoryperiod of the nerve, was deducted from thelatency in each case. This second value was usedto calculate the difference between the maximumand minimum conduction velocities. The meanof the differences between maximum and mini-mum conduction velocities was found for eachgroup; for the alcoholic group the mean was23-75 m/sec, and for the control group 13-63m/sec. This discrepancy between the two groupsis highly significant (P <0-001), and indicatespathological change in the smaller calibre motornerve fibres in the alcoholic patients.

331by copyright.


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from



TABLE 5ULNAR NERVE CONDUCTION VELOCITIES IN PATIENTS AND CONTROLS

Patients Controls

Age oJ' Maximum Minimunm Mininmum Difference Age of' Maximum Minimum Minimum Differencepatient velocity velocity velocity patient velocity velocity velocity(yr) (m/sec.) (m/sec.) if 1 msec. (yr) (n/sec.) (m/sec.) if 1 msec.

allowed for allowed,forrefractory refractoryperiod period(m/sec) (mlsec)

27 49 22 24 25 34 54 30 34 2045 56 25 28 28 44 52 38 42 1042 60 33 37 23 47 49 33 38 1 145 53 3 1 35 1 8 59 55 35 39 1 643 54 27 30 24 37 53 27 3 1 2232 58 35 41 17 26 55 37 41 1429 56 26 29 27 56 49 34 38 1 145 63 3 1 35 28 40 54 34 38 1 6

Patients: mean of the difference between maximum and minimum conduction velocities (I msec allowed for RP)= 23-75 in/sec. SD 3 99.Controls: mean of the difference between maximum and minimum conduction velocities (I msec allowed for RP)= 13-63 rn/sec. SD 4 09.

P < 0-001 (Mann-Whitney test of significance).

TABLE 6CUTANEOUS SENSORY NERVE POTENTIALS IN PATIENTS AND CONTROLS

Age (yr) Cutaneous sensory nerve potentials Sensory nerve mnaximum conductioni H-reflex Ankle jerkvelocities

I III V I III V

Control group < 50 years36-6±6-9 128±43 8-1±3 49±2 1 40-4±5-6 45-9±3-6 47 1±4 1 29-8±2-4 32-4±3-6

Group II. No neuropathy < 50 years38-6 6-3 6-3 ± 3-4 4-4 ± 2-5 1*7 ± 2-1 40-2±- 4-7 46-3 ± 4-7 42-9 ± 3-1 33-1 ± 2-9 38-5 ± 4-5

Group L. With neuropathlv51-3±3 4±1-8 2-3±1-9 0-83±0-7 35-5±3-5 38±2-1 35 ± 3 36-6±2-5 44

All mean values for amplitude of sensory nerve potentials expressed in microvolts ± one S.D.All sensory nerve maximum conduction velocities expressed in metres per second ± one S.D.Latencies of H-reflex and ankle jerk measured in msecs ± one S.D.

CUTANEOUS SENSORY NERVE POTENTIALS, H-REFLEX,AND ANKLE JERK LATENCIES The mean valuesfor the amplitude of the cutaneous sensory nerve

potentials and their maximum velocities forpatients in group I, group II patients aged lessthan 50 years (17 in number), and control sub-jects under 50 years (11 in number) are sum-

marized in Table 6. Although these measure-

ments were made on all five digits, results are

given for digits 1, 3, and 5 only, as these are

entirely representative of the overall findings.These measurements made on group lI patientsand control subjects over 50 years have not beenincluded because it was felt that the numbers ineach group were too few for effective comparisonto be made. Table 6 also includes the latenciesof the H-reflex and ankle jerks for each group.

As is evident, the control group (mean age36f6 + 69 years) is comparable in age with thatsubgroup of alcoholics without neuropathy agedless than 50 years (mean age 38-6 + 6-3 years).The mean values for the sensory nerve maximumconduction velocities in both these groups arevery similar and are within normal limits.There is, however, a small reduction of themaximum sensory nerve velocities in thosepatients who also showed signs of neuropathy.

Alcoholic patients in group I and group IIshow marked reductions in the mean values ofthe amplitude of the cutaneous sensory nervepotentials compared with the normal group.Also, in the group of normal subjects, sensorypotentials were in all cases elicited withoutdifficulty; in the group of alcoholics without

332by copyright.


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from



neuropathy, one patient had no recordablesensory nerve potentials; three other patients inthis group had no recordable sensory potentialsin at least two out of five digits.The average latency of the H-wave for the

normal group was 29X8 msec, and that of theankle jerk, 32X4 msec. The average latency ofboth these reflexes was prolonged in thosealcoholics without neuropathy and increasinglylengthened in those alcoholic patients withneuropathy. Only one patient in group I had anankle jerk which was unimpaired; the latency ofthis was markedly lengthened (44 msec).

DISCUSSION

Our findings indicate that, in both alcoholicswith clinical evidence of polyneuropathy andin those who as yet showed no signs ofperipheral nerve disease, maximum motor con-duction velocities were within normal limitsfor the majority of our patients. Only fivepatients, out of a total of 30 studied, showedlowered conduction velocities in certain segmentsof the peripheral nerves examined. Our results,therefore, differ from those of some workers(Jurko et al., 1964; Mawdsley and Mayer, 1965;Bergamini et al., 1965; Wanamaker and Skill-man, 1966), who found a small reduction inmaximum conduction velocities in a high pro-portion of the alcoholics they examined, but arecompatible with the findings of others (Perfettiet al., 1964; Juul-Jensen and Mayer, 1966;Vacek et al., 1969).

Estimation of minimum conduction velocityin the ulnar nerve of eight alcoholic patientswithout neuropathy revealed impaired rates ofconduction in the smaller calibre motor nervefibres as compared with a similar group ofhealthy control subjects. Perfetti et al. (1967)demonstrated similar involvement of the finermotor nerve fibres in alcoholic patients, most ofwhom however had clinical evidence of poly-neuropathy. It seems, therefore, that damage tofiner motor nerve fibres is detectable beforeclinical symptoms and signs of nerve disorderappear.We have found a marked reduction in

amplitude of the cutaneous sensory nervepotentials in alcoholics, with a small reduction ofthe maximum sensory nerve velocities in thosepatients who also showed clinical signs ofneuropathy. We would, therefore, agree withthe observation of Bergamini et al. (1965) that

sensory nerve fibres are more severely affectedthan motor fibres. This reduction in amplitudeof the sensory potentials is most probably due toactual loss ofdistal sensory nerve fibres, and as themaximum velocities in the digital nerves remainunaffected in alcoholics without neuropathy, thelarger sensory nerve fibres must undergoWallerian degeneration at this stage.The elegant work of Buchthal and Rosenfalck

(1966) on the recovery of sensory nerves afterlocal anaesthesia, showed that normal sensorythresholds of touch and two-point discrimina-tion were reached when the sensory actionpotentials had attained just 4000 of their pre-anaesthetic amplitude. Their findings help toexplain why all modalities of sensation werepreserved in the majority of our patients,although sensory action potentials were reducedby as much as 5000 or more from normal levels.

Prolongation of the latencies of the H-reflexand the ankle jerk occurred in both alcoholicswith and without neuropathy and may be theresult of a number of processes. As function ofthe large efferent fibres was unimpaired, in-volvement of the large afferent (Ia) fibres whichmediate the H-reflex may be assumed. Thedelayed latency of the ankle jerk could be theresult of either involvement of afferent fibres oraffection of the finer motor nerve fibres whichsupply the muscle spindles, thereby renderingthem less sensitive to activation by stretching.Mawdsley and Mayer have speculated on the

nature of the underlying pathological process inalcoholic peripheral nerve disease. They claimedthat the small reductions in maximum conduc-tion velocities in both motor and sensory nerves,which they were able to demonstrate in theirpatients, were indicative of segmental demyelina-tion being the significant pathological process.Our findings strongly suggest that, as far as theterminal portions of the sensory nerve fibres areconcerned, axonal degeneration appears to bethe main pathological event. The detailed histo-logical study of Coers and Hildebrand (1965)confirms this view; in addition to early evidenceof motor end-plate abnormalities, they founddistal degeneration and increased collateralramification of motor nerve fibres to be present,both in alcoholic patients with minimal or noneuropathy. The normal terminal delays wehave found in our patients indicate that,although end-plate abnormalities may exist,they do not interfere with function at nerveendings.

333by copyright.


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from



It is possible, however, that differing patho-logical changes may take place along the courseof a peripheral nerve, and indeed that sensoryand motor nerve fibres may show dissimilarpathological responses to this particular bio-chemical insult. Segmental demyelination maytherefore occur in large sensory fibres causingprolonged H-reflex latencies, but leave, at thatstage anyhow, the large motor fibres un-scathed.The aetiology of alcoholic polyneuropathy

remains uncertain. Shattuck (1928) first sug-gested its nutritional origin, indicating that itwas comparable with beriberi. Strauss (1935)produced evidence against the direct neurotoxiceffect of alcohol and concluded that dietarydeficiencies were important in the developmentof alcoholic polyneuropathy. Denny-Brown(1958) drew attention to the differences betweenthe clinical findings in beriberi and in alcoholicneuropathy: the former is characterized bysymmetrical foot and wrist drop, associated withmuscle tenderness, and only a mild disturbanceof general sensation over characteristic areas;the latter he described as a chronic sensoryneuropathy with marked disturbance of painsensation and often prominent burning paraes-thesiae in the feet. The electrophysiologicalabnormalities we have demonstrated reflect wellthe clinical picture of alcoholic neuropathywhich he outlined.

Evidence of thiamine deficiency was presentin four out of 13 of our patients and deficiencyof this vitamin may therefore have contributedto the peripheral nerve damage we detected. Itis clear, however, that we cannot exclude thepart played by deficiencies of other essentialnutrients in the development of alcoholicperipheral nerve disease.

Warm thanks are expressed to: Dr. F. J. J.Letemendia for his permission to study his patients;Mr. G. Higgins for his biochemical help; MissGloria Woodward and Miss Sylvia Davis for theirelectrophysiological assistance; Miss Diana Staples

for statistical advice; Mr. J. Stan&k for help withtranslation.

This research was carried out with the aid of agrant from the Medical Council on Alcoholism, towhom our thanks are due.

REFERENCES

Bergamini, L., Gandiglio, G., Fra, L., Bergamasco, B.,Bram, S., and Mombelli, A. M. (1965). Alterazioni dellaconduzione nervosa sensitiva e motoria in alcoolisticronici privi di segni clinici di neuropatia periferica.Rivista di Patologia Nervosae Mentale, 86, 31-49.

Buchthal, F., and Rosenfalck, A. (1966). Evoked actionpotentials and conduction velocity in human sensorynerves. Brain Research, v-viii, 1-122.

Coers, C., and Hildebrand, J. (1965). Latent neuropathy indiabetes and alcoholism. Electromyographic and histo-logical study. Neurology (Minneap.), 15, 19-38.

Denny-Brown, D. (1958). Special problems concerningberiberi. A. The neurological aspects of thiamine deficiency.In Nutritional Disease. Proceedings of a Conference onBeriberi, Endemic Goiter and Hypovitaminosis A, Princeton,N.J., 1958, pp. 35-39. Edited by T. D. Kinney and R. H.Follis, Jr. Federation Proceedings, 17, Suppl. No. 2.

Hopf, H. C. (1962). Untersuchungen uber die Unterschiedein der Leitgeschwindigkeit motorischer Nervenfasernbeim Menschen. Deutsche Zeitschrift far Nervenheilkunde,183, 579-588.

Hopf, H. C. (1963). Electromyographic study on so-calledmononeuritis. Archives of Neurology, 9, 307-312.

Jurko, M. F., Currier, R. D., and Foshee, D. P. (1964).Peripheral nerve changes in chronic alcoholics: a study ofconduction velocity in motor nerves. Journal of Nervousand Mental Disease, 139, 488-490.

Juul-Jensen, P., and Mayer, R. F. (1966). Threshold stimula-tion for nerve conduction studies in man. Archives ofNeurology, 15, 410-419.

Mawdsley, C., and Mayer, R. F. (1965). Nerve conduction inalcoholic polyneuropathy. Brain, 88, 335-356.

Norris, A. H., Shock, N. W., and Wagman, I. H. (1953).Age changes in the maximum conduction velocity ofmotor fibres of human ulnar nerves. Journal of AppliedPhysiology, 5, 589-593.

Perfetti, C. C., Milone, F. F., and Pacchiani, A. (1967).Studio dei valori di dispersione della velocitai di conduzione(VDC) delle fibre motorie negli alcoolisti. Rivista diNeurologia, 37, 453-461.

Shattuck, G. C. (1928). The relation of beri-beri to poly-neuritis from other causes. American Journal of TropicalMedicine, 8, 539-543.

Strauss, M. B. (1935). The aetiology of 'alcoholic' poly-neuritis. American Journal of the Medical Sciences, 189,378-382.

Vacek, J., gtika, L., Gr6fova, E., and Warman, C. (1969).Periferni Neuron u Alkoholismu. Casopis Lekaru Ceskjch,108, 331-336.

Wanamaker, W. M., and Skillman, T. G. (1966). Motornerve conduction in alcoholics. Quarterly Journal ofStudies on Alcohol, 27, 16-22.

334by copyright.


http://jnnp.bmj.com

/J N

eurol Neurosurg P


nloaded from


electrophysiological studies alcoholism - bmjtion in the peripheral sensory nerve fibres in these...

Documents