Journal ofNeurology, Neurosurgery, and Psychiatry, 1977, 40, 1083-1091

Effect of body temperature on visual evoked potentialdelay and visual perception in multiple sclerosisD. REGAN, T. J. MURRAY, AND R. SILVER

From the Department of Psychology and Department of Medicine (Camp Hill Hospital),Dalhousie University, Halifax, Nova Scotia, Canada

SUMMARY Seven multiple sclerosis patients were cooled and four heated, but evoked potentialdelay changed in only five out of 11 experiments. Control limits were set by cooling eight and heat-ing four control subjects. One patient gave anomalous results in that although heating degradedperceptual delay and visual acuity, and depressed the sine wave grating MTF, double-flashresolution was improved. An explanation is proposed in terms of the pattern of axonal demy-elination. The medium frequency flicker evoked potential test seems to be a less reliable meansof monitoring the progress of demyelination in multiple sclerosis patients than is double-flashcampimetry or perceptual delay campimetry, although in some situations the objectivity of theevoked potential test would be advantageous.

Neurophysiological studies of single neuroneshave shown that raising the temperature of normalor partially demyelinated axons slightly increasesconduction velocity until just below the tempera-ture at which axonal conduction fails (the block-ing temperature). Raising the temperature stillfurther produces a marked fall of conductionvelocity. Demyelination can so reduce the block-ing temperature that conduction fails at tempera-tures only slightly above normal body temperature(Davis, 1970; Raminsky, 1973; Schauf and Davis,1974).Such neurophysiological observations on single

axons go some way to explaining why exercise orheating may worsen signs and symptoms in somemultiple sclerosis patients, and may even causenew signs and symptoms to appear (Uhthoff, 1890;Davis, 1966; Namerow, 1971). These singleneurone data can also account for the temporaryimprovement that may accompany cooling(Watson, 1959; Boynton et al., 1959; Symington etal., 1977), and can explain why very small changesin body temperature-for example, 0.1 °F-oreven daily circadian variations (Namerow, 1968;Davis et al., 1973) can be sufficient to producelarge effects in patients with multiple sclerosis. Inthe well known 'hot bath' provocation test theeffects of heating are usefully used to help thediagnosis of multiple sclerosis (Davis, 1966).Address for reprint requests: Dr D. Regan, Department ofPsychologyDalhousie University, Halifax, Nova Scotia, Canada B3H 4JL.Accepted 25 Juone 1977

A quite separate point is that temperature ma-nipulations might provide a means of assessing theeffectiveness of techniques for monitoring theprogress of multiple sclerosis on the grounds that,in some patients, small variations of body tem-perature that are well within physiological limitscan mimic the symptomatic effect both of increas-ing demyelination and of remyelination. We havepreviously reported on double-flash campimetry asa possible means of monitoring the progress ofdemyelination and of detecting possible remyelina-tion (Galvin et al., 1976b). Evoked potential re-cording has recently proved to be a diagnosticallyuseful indication of visual pathology in multiplesclerosis, and this method has the advantage of ob-jectivity (Regan, 1977a; Halliday and McDonald,1977; McDonald and Halliday, 1977). Here wereport on evoked potential recording as a possibleobjective means of monitoring the progress ofdemyelination in multiple sclerosis.

Methods

EVOKED POTENTIAL RECORDINGSubjects viewed a diffusing panel subtending480°X480 from a distance of 330 mm. The panelwas lit uniformly by white light to a luminance of110 cd/M2. The panel's luminance could be con-trolled electronically and was sine wave flickeredabout the mean value (30% modulation depth inmost experiments) by feeding the appropriateelectrical waveform to the driving electronics. One

1083

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

D. Regan, T. J. Murray, and R. Silver

of the subject's eyes was occluded by an eyepatch.Two methods were used to record 'medium

frequency' flicker evoked potentials (EP). The'single stimulation method' was to flicker thestimulus light with a sine wave signal of frequencyF Hz. The steady state EP of frequency F Hz wasrecorded by means of a Fourier analyser and itsamplitude and phase recorded as described inRegan (1966) and Milner et al. (1974). Flickerfrequency F was varied between about 13 Hz and28 Hz. A plot of EP phase versus frequency gavethe EP's 'apparent latency' (Regan, 1966; Milneret al., 1974). A speedier method was used in someexperiments. Three sine waves of different fre-quencies (Fl, F2, and F3 Hz) were added togetherand this waveform flickered the stimulator. ThreeFourier analysers, locked to frequencies Fl, F2,and F3 Hz respectively, simultaneously extractedthree EPs from a single electrode derivation anddisplayed the amplitudes and phases of the threeEPs in polar coordinates (Regan, 1976). This'simultaneous stimulation method' directly sampledthe slope of the phase versus frequency plot andthus measured EP latency with greater precisionthan the single stimulation method, since estimateswere less affected by EP variability and non-stationarity (Regan, 1976).

MEASUREMENT OF PERCEPTUAL DELAYPerceptual delay was measured using the pro-cedure described previously (Heron et al., 1974;Regan et al., 1976a). The left eye viewed a small(0.30 diameter solid-state red lamp and the righteye viewed a similar lamp. One lamp was placedslightly above the other, and both fell on thecentral fovea. One lamp was switched on a littlelater (t ms) than the other, and the value of t wasadjusted so that the lamps appeared to switch onsimultaneously. This value t ms then gave per-ceptual delay for one fovea relative to the otherfovea.

MEASUREMENT OF DOUBLE-FLASH RESOLUTIONDouble-flash resolution was measured using theprocedure described previously (Regan, 1972b;Galvin et al., 1976a, b). Subjects viewed a singlesmall red lamp (0.30 diameter) from a distance ofone metre. The background was a diffuse whitesurface illuminated to a luminance of roughly 50cd/M2. The lamp delivered a pair of brief (10 ms)flashes separated by a preset interval. Using themethod of ascending and descending limits theleast interval between the flashes was found forwhich the two flashes could be seen as double. Thefovea of each eye was tested separately. The un-used eye was occluded during tests.

MEASUREMENT OF MODULATION TRANSFER FUNCTIONUSING A SINE WAVE GRATING STIMULUSStandard tests for visual acuity-for example, theSnellen test-assess visual sensitivity for fine detailonly. Visual sensitivity for coarse and mediumdetail as well as fine detail can be tested by usinga sine wave grating stimulus whose spatial con-trast is varied (a sine wave grating looks like ablurred grating of bright and dim bars). Ourmethod was conventional and has been describedelsewhere (Regan et al., 1977a). Sine wave gratingstimuli were displayed on a cathode ray oscillo-scope by electronic means. Spatial frequency wasdefined as the number of (bright plus dim) barsper degree of visual angle, and this could bevaried. Contrast was defined as Imax-Im1n, where

Imax+ IminImax and Imin were the luminances of a bright anddim bar respectively. For each given spatial fre-quency the contrast was reduced so as to find thecontrast for which the grating could just be seen.Thus, the contrast threshold was a measure ofvisual sensitivity for each of the spatial frequenciestested. The stimulus field subtended 3.5° X4.5°(horizontally), was of mean luminance 17 cd/M2,and had a white surround of luminance 11 cd/M2subtending about 210 X 140 (horizontally).

COOLING PROCEDURESubjects ingested 400 ml of finely crushed ice,within five to 10 minutes, either unflavoured orflavoured with orange juice, as in Galvin et al.(1976b).

WARMING PROCEDUREBoth legs were immersed to upper shin level inwarm water at about 440C, and the subject waswrapped in blankets with two hot water bottlesaround the abdomen. Warming was continueduntil the patient became almost intolerably hotor until the patient reported markedly degradedvision: this took on average roughly 30 min.

Control subjects and patients

Control subjects were matched for age to indi-vidual patients. The diagnostic criteria for multiplesclerosis were those set out by Schumacher et al.(1968). Patients were classified into clinicallydefinite, probable, and possible groups accordingto the scheme described by Rose et al. (1976). Inaddition, each patient was rated on the 10 pointdisability scale of Kurtzke (1965). A total of 10multiple sclerosis patients were tested in this EPstudy. Clinical details are summarised in the Table.

1084

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

Effect of body temperature on visual evoked potential delay and visual perception

Table Clinical summary and results for the 10 patients tested by evoked potential recording

Paticnt History Diagnosis History Optic discs Field Comments Result of Result ofRBN of temperature defect heating cooling

sensitivity

I - definite - pale discs - spinal form of MS2 - definite -- normal - spinal form of MS +3 - definite - left disc pale spinal and brainstem with +

some cerebral signs4 - definite - - - spinal form of MS5 - definite - - - spinal-cerebellar type +6 - definite - - - spinal: blurred vision

in left eye, but thateye has cataract

7 - definite -,- ---- - 'washed out vision'cerebellar signs

8 ± definite +-4- + - spinal and brainstem signs +9 - definite - - - spinal form10 - definite - - - spinal-cerebellar- +

brainstem form

RBN =retrobulbar neuritis; MS= multiple sclerosis

Results

The baseline delay for all 10 patients was outsidecontrol limits (P<0.05) in accord with previousfindings (Milner et al., 1974; Regan et al.,1976a, b).Cooling had no noticeable effect on EP delay

(less than 10% change) in the eight control sub-jects who were cooled. The control data in Fig. 1Awere recorded by means of the single frequencymethod, and the simultaneous stimulation methodwas used to obtain the data in Fig. l B (seeMethods).

400-1

1501 A

100-

co 50-E 200-

UzX 150-

' 100-

z 200-

0Q- 150-CLUo 100-

ingest ice

B

ingest ice

C

healt

I. .

ui 6 30 do 90 120TIME min

Fig. 1 Neither heating nor cooling produced anysystematic changes of evoked potential latency incontrol subjects. Plot B was recorded using a fastrecording technique.

300-

Seven multiple sclerosis patients were cooled.All were diagnosed as having definite multiplesclerosis. Two of these patients showed a clear,temporary reduction of EP delay after ingestionof ice (Fig. 2A, B). One patient's EP delay showeda temporary 'rebound' to a value greater thanhis baseline (Fig. 2B). Temporary rebounds havebeen observed previously with visual acuity as ameasure (Michael and Davis, 1973). A thirdpatient showed smaller, but still clear temporaryreductions in EP latency. The data on the remain-

A

ingest iceEl_(5200- Iz I

100

0

w300fB> ingest ice

200{ - rebound

0 30 60 90TIME min

Fig. 2 Cooling reduced evoked potential delay insome multiple sclerosis patients. Other patients didnot show this effect. Plot B shows a temporary'rebound'. A =patient 3; B=patient 2.

1085

1

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

D. Regan, T. J. Murray, and R. Silver

ing four patients could not be distinguished fromcontrol measurements.Heating had no noticeable effect on EP delay

(less than 10% change) for the four control sub-jects who were heated. The control data in Fig. ICwere obtained by means of the single stimulationtechnique (see Methods).Four multiple sclerosis patients were heated.

Two patients (8 and 10) showed an increase oflatency (Fig. 3A, B). Patient 8 (Fig. 3A) reported'very hot, vision darkened' at a, 'see better' at b,and 'back on normal temperature' at c. The re-maining two patients' data could not be dis-tinguished from control measurements (forexample, Fig. 3C).These EP findings contrast with the results of

previous exploratory experiments on the effect oftemperature changes upon the delay of visualperception (Regan, Milner, and Heron, unpub-lished observations). Patients and control subjectswore underclothing which incorporated coils ofplastic tubing through which cold water was cir-culated. The difference in perceptual delays be-tween the left and right eyes was measured asdescribed previously, using a light stimulus of0.3° subtense (Regan, 1972b; Heron et al., 1974;Regan et al., 1976b). Measurements were carriedout: (a) before cooling; (b) during cooling; (c) onehour after end of cooling. Delays for three controlsubjects were (in ms): (a) 11, (b) 15, (c) 8; (a) 14,

250-

200-

c,,250E

0200-zLu

< 150--J

z- 1 00-z

250-

o

01

0

ui150J

A

heat

B

heat

C

heat

0 30 60 90 120TIME min

Fig. 3 Heating increased evoked potential delay insome multiple sclerosis patients (A, B), but had no

effect in others (C). A =patient 8; B=patient 10;C=patient 9.

(b) 25, (c) 4; (a) 11, (b) 11, (c) 2. In contrast,cooling caused a clear change of perceptual delayin all seven patients with unilateral retrobulbarneuritis who were tested. Results were as follows:patient 1: (a) 132, (b) 24, (c) 14; patient 2: (a) 35,(b) 6, (c) 0; patient 3: (a) 41, (b) 36, (c) 11; patient4: (a) 194, (b) 138, (c) 105; patient 5: (a) 23,(b) -7, (c) 2; patient 6: (a) 94, (b) 15, (c) 18.Clearly, delays did not return to precooling valueswithin one hour; but measurements continued forfour hours after the end of cooling in one patient(6) showed that a steady value was reached aftertwo hours. A seventh patient was both warmed(twice) and cooled (twice). Baseline values fordelay, measured over a two-day period were 126,156, 165, 157, 216, and 127 ms (mean 158). Delaysduring the two coolings were 105 ms and 90 msand during the two heatings 195 ms and 180 ms.One patient who showed no EP effect reported

that vision in her left (affected) eye deterioratednoticeably compared with her right eye during thetest. We, therefore, assessed her visual acuitythoroughly before, during, and after heating byusing a sine wave grating stimulus in the waydescribed elsewhere (Murray et al., 1977; Reganet al., 1977). The sine wave grating test enablesvisual sensitivity to be assessed for coarse,medium, and fine detail, rather than only for finedetail as in conventional tests such as the Snellentest. The results, shown in Fig 4, confirmed andextended the patient's informal report. Figure 4plots visual sensitivity (that is, contrast sensitivity)versus spatial frequency (that is, the number ofbright bars per degree of visual angle for thegrating stimulus). The left eye had previouslyexperienced an attack of retrobulbar neuritis.Preheating data for this eye are marked 1 in Fig.4A. Plot 2 was recorded during heating and showsa depression of visual contrast sensitivity for allspatial frequencies. Plots 1 and 2 also show thatthe highest (cutoff) spatial frequency that couldbe seen with the left eye was also depressed, andthis is equivalent to saying that the Snellen acuitywould have been depressed. A general depressionof visual contrast sensitivity has been reportedpreviously in multiple sclerosis, though as a lastingrather than temporary condition (Regan et al.,1977). Plot 3 was recorded 30 min after heatingended and shows a 'rebound' to a visual sensitivityhigher than before heating. Plot 4 shows that 60minutes after ending heating the curve was re-turning to baseline.

Corresponding plots for the right ('unaffected')eye are shown in Fig. 4B. Heating had little effecton the cutoff spatial frequency (equivalent toSnellen acuity). Unexpectedly, it seems that when

1086

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

Eflect of body temperature on visual evoked potential delay and visual perception

Aleft eyeRBN

heat i2

i 2lb10SPATIAL FREQUENCY c/deg

right eye

2

1rrecover

3

4

05 i 2 5 10

SPATIAL FREQUENCY cideg

Fig. 4 Effect of heating on visual acuity andcontrast sensitivity in patient 7. The left eye hcpreviously experienced an attack of retrobulbaneuritis. The preheating curve (1) was shifteddownwards during heating (2), showing that visensitivity was degraded for all spatial frequenVisual sensitivity showed an 'overshoot' 30 miuheating was ended (3), but was returning to ba60 min after heating was ended (4). Visual acu

cutoff spatial frequency) behaved similarly. Viacuity in the right ('unaffected') eye showed litchange. Surprisingly, there was some suggestiothe right eye's sensitivity to medium spatialfrequencies increased at about the same time (

the left eye's decreased.

the left eye's sensitivity was depressed bythe right eye's sensitivity to middle rangefrequencies was enhanced, finally returningbaseline levels 60 minutes after the end c

ing. Depressing the contrast sensitivity ofsubjects by adapting to some value of spalquency has been reported not to enhanc(tivity at neighbouring spatial frequencieset al., 1976), although enhancement atspatial frequencies has been reported (De1977).We measured the delay of visual percep

this patient by a method described pre(Heron et al., 1974; Regan et al., 1976a, b).5 plots the difference between the perceptua

E+20-- heat

baseline/_ < 20 *~~~-----------------------------------------------------------------------2 <-420-

c -60-

Q- 6 6b 180i6TIME min

Fig. 5 Effect of heating on the delay of visual20 perception in patient 7. Perceptual delay is the

difference between left and right fovea: negativevalues mean the left was slower than the right.Heating increased the delay of the left ('affected') eyecompared with the right eye.

for the left and right foveae. Heating increased*5 the delay of the left eye relative to the right eye,

and after the cessation of heating there was a@ 'rebound'. Thus, perceptual delay, but not EP

delay, was affected by heating in this patient.We then measured double-flash resolution in

this patient during and after heating. Figure 6shows the surprising results. Baseline values of

20 double-flash threshold were outside normal limitsfor both eyes (P<0.01), but heating improvedpatial double-flash resolution. This behaviour is the op-

dr posite to our previous observations: double-flashbodily resolution was markedly degraded by heating inisual all patients previously tested (Galvin et al., 1976b).cies. Note that temperature changes had no effect inz after control subjects.selineity (the

'sualrtleIn that

2) as

heating,spatialto nearf heat-controltial fre-e sensi-Barlowremote-Valois,

)tion inviouslyFigure

I delays

1201

Q)E 110z

100

0 90-U)

w

a: 80I

Cl)

< 70oU-

W 60-m

o 50-,>I

Q

heat0

.0

.

, .,

'I

0 30 60 90 120 150 180 21(TIME min

Fig. 6 Anomalous effect of heating on double-flashperception in patient 7. Ordinates plot the timeinterval between the onset of two brief flashes forwhich they are just seen as double rather than single.

=left ('affected') eye, .... =right eye.

Heating improved double-flash resolution for botheyes. All multiple sclerosis patients previously testedshowed the opposite effect, and all control subjectsshowed no effect.

0

0*5

1-

q0z 50wLC

CO)CCz DU0o0100

0-5

B0-55-

0

° 5-Ie)wac 10-I

(n

Cc 50z01000

1087

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

D. Regan, T. J. Murray, and R. Silver

Discussion

The visual evoked potential is not a single entity.Among the many different types of visual EP areresponses to pattern reversal, responses to flash-ing a spatially unpatterned light, responses toflickering a spatially unpatterned light at fre-quencies between 13 and 25 Hz ('medium fre-quency' EPs), and 'high frequency' flicker EPswhose frequencies lie between about 40 and 60Hz (Regan, 1972a, 1975). There is evidence thatof these various types of visual EP several are notonly generated in different (though overlapping)areas of cortex, but also reflect the processing ofdifferent visual signals in parallel channels at bothperipheral and central levels (Regan, 1972a, 1973,1975).

All these types of EP have been recorded inmultiple sclerosis patients with the aim of findinga diagnostically useful indicator of visual path-ology. Evoked potentials to pattern reversal oflarge (50 min arc) checks are delayed in multiplesclerosis (Halliday et al., 1972, 1973; Arden, 1973;Milner et al., 1974; Assleman et al., 1975; Mc-Donald and Halliday, 1977). Evoked potentials tosmall checks are less reliably delayed (Milner etal., 1974). Evoked potentials to spatially unpat-terned flash also seem to be a less reliable indicatorof multiple sclerosis than large check EPs (Richeyet al., 1971; Halliday et al., 1972), although thisview has been challenged (Feinsod and Hoyt,1975). Medium frequency flicker EPs are reliablydelayed in multiple sclerosis (Milner et al., 1974;Regan et al., 1976a). (Figure 3 in Regan (1977b)provides evidence that flicker EPs resemble EPs tolarge checks or low spatial frequencies, and differfrom EPs to small checks or high spatial frequen-cies.) Three important features of the EP test inmultiple sclerosis patients are that some specifictypes of visual EP show a high incidence of abnor-mality in multiple sclerosis even when there areno corresponding abnormal clinical signs, that theEP test can demonstrate lasting visual damageeven after clinical recovery of vision and that it isan objective test (McDonald and Halliday, 1977;Regan, 1977a). Note, however, that although EPsare not delayed by retinovascular lesions (Milneret al., 1974), multiple sclerosis is not the onlycause of delay (Cappin and Nissim, 1975; Halli-day et al., 1976; McDonald and Halliday, 1977)

All 10 patients in our present study had delayedEPs. Seven patients were cooled and four heated,but EP latency changed in only five out of 11experiments: in one case (patient 7) latency wasmeasurably constant to a precision of 6%, thoughvisual changes were noticeable to the patient.

On the other hand, in previous preliminary re-search, perceptual delay was reduced by coolingin eight out of eight experiments on seven patients,and increased by heating in two out of two experi-ments. Thus, the effect of body temperature onEP delay seems less consistent than the effects oftemperature on perceptual delay.

Double-flash campimetry is a psychological (per-ceptual) test that can not only detect subclinicalvisual pathology in multiple sclerosis (in nine outof 11 spinal patients [Galvin et al., 1977]); it canalso demonstrate lasting visual damage after acuteretrobulbar neuritis (13 out of 14 patients), evenwhen visual acuity has recovered clinically (Galvinet al., 1976a), and can detect the progression of thedisease (in 12 cases of advanced multiple sclerosis[Galvin et al., 1977]). In a previous study on fourmultiple sclerosis patients three were heated andthree cooled using similar procedures to ours(Galvin et al., 1976b). Although only one of the10 patients had previously noticed any visualeffects produced by temperature changes, clearalterations in double-flash resolution were pro-duced in all six temperature experiments. Thus EPdelay seems to be less sensitive to temperaturechanges than is double-flash perception. However,although our paradoxical findings with patient 7may be unusual, they should be kept in mind whenusing the double-flash test.

Questions could be raised as to the compara-bility of the results of the various tests. It is, ofcourse, possible that we would have found EPdelay to have been affected by temperature in allpatients had we heated and cooled the patientsmore strongly. Our point here, however, is todiscuss whether the EP test is less sensitive totemperature changes than are the other tests (and,therefore, presumably less sensitive to changes indemyelination). Since we used the same methodsas in the Galvin et al. (1976b) double-flash studyit seems unlikely that the heating and cooling pro-cedures were less effective in the present EP study.Although we used a different patient group, theirclinical descriptions were not dissimilar. In com-paring the results of the EP study with the per-ceptual delay data reported here it should be keptin mind that heating and cooling procedures weredifferent. On the other hand it seems unlikely thatheating was less effective in our EP experiments,as all patients were heated close to their limitsof toleration. At any rate, for patient 7 in thepresent study we used similar heating proceduresin all tests.

Differences between the effect of temperatureon the EP test and the perceptual tests seem un-likely to be due to the fact that the EP test

1088

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

Effect of body temperature on visual evoked potential delay and visual perception

samples a much larger retinal area than either thedouble-flash or delay tests (480 compared with0.30 diameter). Nevertheless, because of the smal.size of the test field the perceptual tests can inprinciple provide more information than EP testsabout the progress of demyelination. By meansof visual field plots, double-flash campimetry, anddelay campimetry can indicate not only changesin the sizes of 'islands' of demyelination, but alsochanges in the local severity of the defect (Galvinet al., 1976a, b, 1977).Although the rule seems to be that (if tempera-

ture has any effect at all) heating worsens andcooling improves the signs and symptoms of mul-tiple sclerosis, there are a few reports that signsand symptoms may worsen on exposure to cold(Simons, 1937; Watson, 1959; Geller, 1974). Al-though, as Geller (1974) pointed out, it is possiblethat in some of these cases shivering might havecaused body temperature to rise on exposure tocold, it nevertheless seems possible that a genuineparadoxical response to temperature changes canoccur. A case in point is that heating improveddouble-flash perception for patient 7, contrary toexperience with other patients. This paradoxicalresponse to heating was peculiar to double-flash,since heating reduced visual acuity and increasedperceptual delay. Galvin et al.'s (1976b) findingthat the double-flash perception of multiplesclerosis patients was degraded by heating can beunderstood if heating brings axons closer to theirblocking temperatures, and thus reduces the maxi-mum firing frequency (Regan, 1972b; Regan et al.,1976b): an alternative explanation is that ad-jacent axons have different delays, so that visualsignals from different parts of the stimulatedretinal area are dispersed in time (McDonald,1974). Neither explanation can account for theimprovement in double-flash perception on heat-ing shown in Fig. 6. Perhaps this anomalous im-provement in double-flash perception might beexplained if the following unusual situation ob-tained in this patient. At normal body temperaturemost fibres were far from their blocking tempera-ture, but conduction was close to failing in somescattered fibres. Conduction along those latterfibres would increase the net temporal dispersionof visual signals so that their selective blockagemight improve double-flash resolution. Heatingwould have little effect on conduction speed forthe majority of axons (thus explaining the EPdata), except in a few localised regions of moresevere demyelination (thus explaining the per-ceptual delay data). Interestingly, this patientvolunteered the report that the perceptual delaytest was easier to do during heating than either

before or after: presumably simultaneity wasmore clearly defined.

In this report we explore whether recordingmedium frequency flicker EPs might provide anobjective means of monitoring the progress of de-myelination and possibly also of monitoring experi-mental therapy. We find that, although this EPtest provides a reliable indication of visual path-ology, EP delay seems to be less affected bychanges in body temperature than is either double-flash perception or perceptual delay. We concludethat the medium frequency flicker EP test seemsto be a less sensitive and possibly less reliablemeans of monitoring the progress of demyelina-tion in multiple sclerosis patients than is double-flash campimetry or perceptual delay campimetry.However, the objective nature of the EP testwould be an advantage in some situations.

We are very grateful to the patients and controlsubjects who generously assisted in this study. Wethank Dr D. B. King for allowing us to examinesome of his patients and for supplying diagnosticassessment of them. We are grateful to AnnetteMcCaul for assistance in testing patients. Wethank Nancy Beattie for assistance in preparingthis manuscript. This work was supported by theMultiple Sclerosis Society of Canada, the MultipleSclerosis Society of Great Britain and NorthernIreland, and the Medical Research Council (UK).RS was supported as a Summer Student by theDepartment of Medicine, Camp Hill Hospital andby the Canadian Medical Research Council.Wilkinson Sword Research developed and suppliedthe double-flash and delay campimeter.

References

Arden, G. B. (1973). The visual evoked response inophthalmology, Proceedings of the Royal Societyof Medicine, 66, 1037-1044.

Assleman, P., Chadwick, D. W., and Marsden, C. D.(1975). Visual evoked responses in the diagnosisand management of patients suspected of multiplesclerosis. Brain, 98, 261-282.

Barlow, H. B., MacLeod, D. I. A., and van Meeteren,A. (1976). Adaptation to gratings: no compensatoryadvantages found. Vision Research, 16, 1043-1046.

Boynton, B. L., Garramone, P. M., and Buca, J. T.(1959). Observations on the effect of cool baths forpatients with multiple sclerosis. Physical TherapyReview, 39, 297-299.

Cappin, J. M., and Nissim, S. (1975). Visual evokedresponses in the assessment of field defects inglaucoma. Archives of Ophthalmology, 93, 9-18.

Davis, F. A. (1966). Hot-bath test in the diagnosis ofmultiple sclerosis. Journal of the Mount SinaiHospital, 33, 280-282.

E

1089

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

1090

Davis, F. A. (1970). Axonal conduction studies basedon some considerations of temperature effects inmultiple sclerosis. Electroencephalography andClinical Neurophysiology, 28, 281-286.

Davis, F. A., Michael, J. A., and Tomaszewski, J. S.(1973). Fluctuations of motor function in multiplesclerosis related to circadian temperature variations.Diseases of the Nervous System, 34, 33-36.

DeValois, K. K. (1977). Spatial frequency channelsrevisited: evidence for mutual inhibition. Universityof Houston Symposium. Abstracts.

Feinsod, M., and Hoyt, W. F. (1975). Subclinicaloptic neuropathy in multiple sclerosis. Journal ofNeurology, Neurosurgery, and Psychiatry, 38, 1109-1114.

Galvin, R. J., Regan, D., and Heron, J. R. (1976a).Impaired temporal resolution of vision after acuteretrobulbar neuritis. Brain, 99, 255-268.

Galvin, R. J., Regan, D., and Heron, J. R. (1976b).A possible means of monitoring the progress ofdemyelination in multiple sclerosis: effect of bodytemperature on visual perception of double lightflashes. Journal of Neurology, Neurosurgery, andPsychiatry, 39, 861-865.

Galvin, R. J., Heron, J. R., and Regan, D. (1977).Subclinical optic neuropathy in multiple sclerosis.Archives of Neurology (Chicago). In press.

Geller, M. (1974). Appearance of signs and symptomsof multiple sclerosis in response to cold. Journal ofthe Mount Sinai Hospital, 41, 127-130.

Halliday, A. M., and McDonald, W. I. (1977). Patho-physiology of demyelinating disease. British MedicalBulletin, 33(1), 21-27.

Halliday, A. M., McDonald, W. I., and Mushin, J.(1972). Delayed pattern-evoked responses in opticneuritis. Lancet, 1, 982-985.

Halliday, A. M., McDonald, W. I., and Mushin, J.(1973). Delayed pattern-evoked responses in opticneuritis in relation to visual acuity. Transactions ofthe Ophthalmological Society of the United King-dom, 93, 315-324.

Halliday, A. M., Halliday, E., Kriss, A., McDonald,W. I., and Mushin, J. (1976). The pattern evokedpotential in compression of the anterior visual path-ways. Brain, 99, 357-374.

Heron, J. R., Regan, D., and Milner, B. A. (1974).Delay in visual perception in unilateral opticatrophy after retrobulbar neuritis. Brain, 97, 69-78.

Kurtzke, J. F. (1965). On the evaluation of disabilityin multiple sclerosis. Neurology (Minneapolis), 15,654-693.

McDonald, W. I. (1974). Pathophysiology in multiplesclerosis. Brain, 97, 179-196.

McDonald, W. I., and Halliday, A. M. (1977). Diag-nosis and classification of multiple sclerosis. BritishMedical Bulletin, 33(1), 4-19.

Michael, J. A., and Davis, F. A. (1973). Effects of in-duced hyperthermia in multiple sclerosis: differencesin visual acuity during heating and recovery phases.Acta Neurologica Scandinavica, 49, 141-151.

Milner, B. A., Regan, D., and Heron, J. R. (1974).Differential diagnosis of multiple sclerosis by visual

D. Regan, T. J. Murray, and R. Silver

evoked potential recording. Brain, 97, 755-772.Murray, T. J., Regan, D., and Silver, R. (1977).Hidden visual loss in multiple sclerosis. Neurology(Minneapolis), 27, 345.

Namerow, N. S. (1968). Circadian temperature rhythmand vision in multiple sclerosis. Neurology (Minne-apolis), 18, 417-422.

Namerow, N. S. (1971). Temperature effect on criticalflicker fusion in multiple sclerosis. Archives ofNeurology (Chicago), 25, 269-275.

Raminsky, M. (1973). The effects of temperature onconduction in demyelinated nerve fibers. Archivesof Neurology (Chicago), 28, 287-292.

Regan, D. (1966). Some characteristics of averagesteady-state and transient responses evoked bymodulated light. Electroencephalography and Clini-cal Neurophysiology, 20, 238-248.

Regan, D. (1972a). Evoked potentials in psychology,sensory physiology and clinical medicine. Chapmanand Hall: London, John Wiley: New York.

Regan, D. (1972b). UK Patent Application No. 4865/72. US Patent No. 3, 837, 734. W. German No. 2,304, 808. Wilkinson Sword Ltd.

Regan, D. (1973). Parallel and sequential processingof visual information in man: investigation byevoked potential recording. In Photophysiology,Vol. 8, pp. 185-208. Academic Press: New York.

Regan, D. (1975). Recent advances in electrical re-cording from the brain (Review). Nature, 253, 401-407.

Regan, D. (1976). Latencies of evoked potentials toflicker and to pattern speedily estimated by simul-taneous stimulation method. Electroencephalog-raphy and Clinical Neurophysiology, 40, 654-660.

Regan, D. (1977a). Visual tests in multiple sclerosis.Proceedings of the San Diego Biomedical Sym-posium, 16, 87-95.

Regan, D. (1977b). Assessment of visual acuity byevoked potential recording: ambiguity caused bytemporal dependence of spatial frequency selectivity.Vision Research. In press.

Regan, D., Milner, B. A., and Heron, J. R. (1976a).Delayed visual perception and delayed visual evokedpotentials in the spinal form of multiple sclerosisand in retrobulbar neuritis. Brain, 99, 43-66.

Regan, D., Silver, R., and Murray, T. J. (1977).Visual acuity and contrast sensitivity in multiplesclerosis: hidden visual loss. Brain, 100, 563-579.

Regan, D., Varney, P., Purdy, J., and Kraty, N.(1976b). Visual field analyser: assessment of delayan'd temporal resolution of vision. Medical andBiological Engineering, 14, 8-14.

Richey, E. T., Kooi, K. A., and Tourtellotte, W. W.(1971). Visual evoked responses in multiple sclerosis.Journal of Neurology, Neurosurgery, and Psy-chiatry, 34, 275-280.

Rose, A. S., Ellison, G. W., Myers, L. W., andTourtellotte, W. W. (1976). Criteria for the clinicaldiagnosis of multiple sclerosis. Neurology (Minne-apolis), 26, 20-22.

Schauf, C. L., and Davis, F. A. (1974). Impulse con-duction in multiple sclerosis: a theoretical basis for

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from

Efect of body temperature on visual evoked potential delay and visual perception

modification by temperature and pharmacologicalagents. Journal of Neurology, Neurosurgery, andPsychiatry, 37, 152-161.

Schumacher, G. A., Beebe, G., and Kilber, R. F.(1968). Problems of experimental trials of therapyin multiple sclerosis: Report of the panel on evalua-tion of experimental trials in multiple sclerosis.Annals of the New York Academy of Sciences, 122,522-568.

Simons, D. J. (1937). A note on the effects of heatand cold upon certain symptoms of multiplesclerosis. Bulletin of the Neurological Institute ofNew York, 6, 385-386.

Symington, G. R., MacKay, I. R., Currie, I. R., andCurrie, T. T. (1977). Improvement in multiplesclerosis during prolonged induced hypothermia.Neurology (Minneapolis), 27, 302-303.

Uhthoff, W. (1890). Untersuchungen uber die bei dermultiplen herdsklerose vorkommenden augenstorun-gen. Archiv fur Psychiatrie und Nervenkrankheiten,21, 55-116, 303-410.

Watson, C. W. (1959). Effect of lowering body tem-perature on the symptoms and signs of multiplesclerosis. New England Journal of Medicine, 261,1252-1259.

1091

guest. Protected by copyright.

on Decem

ber 30, 2021 byhttp://jnnp.bm

j.com/

J Neurol N

eurosurg Psychiatry: first published as 10.1136/jnnp.40.11.1083 on 1 N

ovember 1977. D

ownloaded from