Ophthal. Physiol. Opi. Vol, 17, No, 3. pp, 232-238. 1997C; 1997 The College of Optometrists, Published by Elsevier Science Ltd

Printed in Great Britain0275-5408/97 $I7.(X) + 0,00

PII: S0275-5408(96)00074-9

Spectral sensitivity in patients with dysthyroideye disease

Sharanjeet-Kaur\ Chris M. Dickinson'Ian J. Murray^

Eammon O'Donoghue^ and

^ Dept of Optometry, University Kebangsaan Malaysia, Jalan Raja Muda, 50300 Kuala Lumpur,Malaysia; Dept of Optometry and Visual Sciences, UMIST, PO Box 88, Manchester M60IOD, UK; and ^Senior Registrar, Manchester Royal Eye Hospital, Oxford Road, ManchesterM l 3 9WH UK

SummaryThe majority of patients with dysthyroid eye disease have an acquired colour vision defect.However, no psychophysical investigation of selective damage to colour or flicker pathwayshas been carried out. In order to clarify the nature of the visual pathology, we have used apsychophysical technique (spectral sensitivity! to selectively stimulate the chromatic andachromatic mechanisms. Spectral spots of size 1 ° presented at a rate of 1 Hz on a brightlOOOtd white background are detected by the chromatic mechanism but a rate of 25 Hzreveals the achromatic mechanism. Fifteen patients (28 eyes) between the ages of 50 -70years were tested. The study showed that all patients had reduced spectral sensitivity, either1 Hz, 25 Hz or both. The patients with reduced 1 Hz or 25 Hz spectral sensitivity only had ashorter systemic and ocular duration of the condition, had no proptosis, normal intraocularpressures in primary gaze, slightly higher intraocular pressures on upgaze, normal visual fieldplots and FM 100 Hue error scores higher than the normal age-matched values. The patientswith reduced both 1 Hz and 25 Hz spectral sensitivities had a longer systemic and ocularduration of the condition, had proptosis, normal intraocular pressures in primary position,higher intraocular pressures on upgaze and higher FM 100 Hue error scores than the age-matched normals and those in Groups 1 and 2, A total of 50% of patients in Group 3 haddefective visual field plots. These data suggest that there is damage of the large achromaticfibres and small chromatic fibres in dysthyroid eye disease. The mechanism of the damagecould be one of ischaemic or mechanical or both. ©1997 The College of Optometrists.Published by Elsevier Science Ltd,

Introduction

Graves' disease is the most common variety of hyper-thyroidism and is regarded as having an autoimmune basis(Kanski and McAllister, 1989). Most palients with Graves'disease have evidence of thyroid overaclivity associatedwith eye signs such as eyelid retraction, infillrative ophthal-mopathy, proptosis, restrictive myopathy and optic neuro-pathy (Kanski and McAllister, 1989).

The prevalence of dysthyroid optic neuropathy is probablyless than 5% in patients with thyroid disease. All the mech-anisms responsible for the condition are not clear. However,it seems likely that it is caused by a direct compression of

Received: 4 April 1995Revised form: 15 October 1996

the Optic nerve or its blood supply at the orbital apex by thecongested and enlarged recti muscles (Kanski and McAllister,1989). CT scans suggest that there is a simple compressionof the optic nerve at the orbital apex within the massivelyoedematous extraocular muscles (Sergott and Glaser, 1981).It has been suggested that this compression probably causesinterruption in the axoplasmic flow (Sergott and Glaser,1981). However, there are no published experimental studieson axoplasmic fiow in cases of optic nerve compression.This compression may lead to severe but preventable visualimpairment. Treatment may be oral corticosteroids. orbitalirradiation or surgical orbital decompression. Visual fieldexamination usually demonstrates a central scotoma with orwithout inferior arcuate bundle defects (Sergott and Glaser,1981; Trobe et al.. 1978), Apart from this, there is animpairment of colour vision (Kanski and McAllister, 1989).

232

Spectral sensitivity in patients with dysthyroid eye disease: Sharanjeet-Kaur et al. 233

Extensive colour vision studies on patients with dysthyroideye disease have not been conducted. Most studies havebeen carried out to look at the effects of different types oftreatment of the eye disease (Brown et a/., 1963; Riley,1972; Gorman et at.. 1974; Ravin et al.. 1975; Trobeet al., 1978; Linberg and Anderson, 1981; Panzo andTomsak, 1983). However, most of these studies do mentionthat there is colour vision dysfunction observed using theAO H-R-R pseudoisochromatic plates or Ishihara colourplates in these patients. Another study using FM lOO-Huetest on patients with optic nerve compression including onepatient with Graves* disease show that there was a generalcolour vision dysfunction without any tendency towards aspecific red-green or blue-yellow confusion (Paulus andPlendl, 1991).

There is also no known psychophysical study about theoptic nerve compression as a consequence of dysthyroideye disease which gives insight into the type of fibresaffected, as is known in glaucoma. Therefore, studyingspectral sensitivity measurements could probably givesome understanding into the type of the functional defect(i.e. whether chromatic or flicker function) most commonlyseen in dysthyroid eye disease.

Material and methods

Apparatus

This study was carried out at the Manchester Royal EyeHospital (MREH). A 2-channel Maxwellian-view apparatuswas used (Figure / ) to conduct the experiments. The

FW

M

Figure 1. Schematic diagram of the apparatus used atMREH. L, and Lj converging lenses; M i , front surfacemirror; FH, filter holder; St,, stop which controls the sizeof the background field; FT, slide containing the annulus;L3, L4, converging lenses; Stj, stop of 2,2 mm diameter(1"); Sh, shutter assembly; FW, filter wheel mounted ona 12-wav switch; L5, converging lens; M j , front surfacemirror, W,, neutral density wheel; BC, glass plate beamcombiner; AL, achromatizing lens; AP, 2 mm exit pupil; LC,lens cell for lenses to correct the subjects' refractive error;SQ, 12 V 50 W quartz iodide light source.

background channel consisted of a 12 V, 50 W quartz iodidelight source Sy, converging lens combination (L, and L,),front surface mirror, M,, filter holder FH, for the neutraldensity filters, a field stop St,, which controlled the size ofthe background field, a slide containing a dark anulus ofwidth 10 min of arc, FT, placed close to St,, and lenscombinations L, and Lj. A lOOOtd white background ofcolour temperature 2600 K was produced using an Ilford3.0 log unit neutral density filter in FH. The colour temp-erature was determined using a PR 1500 photometer. Themeasurement was made through red and blue filters foundin the photometer. The colour temperature was then cal-culated as the ratio of measurement through the blue filterand measurement through the red filter. Retinal illuminationwas determined using the following formula:

T= 31400 X L (Cd/m-) (see Westheimer, 1966)

The luminance L in candela/m' was measured using thePR1500 photometer. It was found that in order to obtainlOOOtd, a 3.0 neutral density filter was needed.

The test channel consisted ofthe same quartz iodide lightsource; a 2.2 mm diameter (1°) stop, St;; shutter assembly,Sh; filter wheel mounted with a 12-way switch, FW; lenssystem, L5 (which formed a parallel beam); front surfacemirror, M,; 5% inch diameter Kodak neutral density wedge(W,) of 0—6 log units, mounted upon a precision potentio-meter and operated manually; glass plate beam combiner,BC; ocular assembly comprising an achromatizing lens(AL) to counteract the effects of chromatic aberrations ofthe eye (Bedford and Wyszecki, 1957; Lewis et al., 1982);lens cell, LC (for lenses to correct the subjects refractiveerror); and a 2 mm exit pupil. AP. The coloured filters forthe spectral light were B40 Balzer interference filters, withpeaks at 402, 423, 450, 474, 497, 527, 554, 574, 601, 622,652 nm and an average bandwidth at half-height of 10 nm.

An analogue computer was also used to correct for non-linearity and non-neutrality of the neutral density wedge,and also to correct the quantum intensity of the test flash ateach wavelength wben the neutral density wedge is set tozero (Zisman era/,, 1977). The data was then plotted by thecomputer as sensitivity in log relative quantal units versuswavelength, in all measurements,, quantal efficiency wastaken into account.

Subjects

Fifteen patients (28 eyes) between the ages of 50-70 years(mean age = 58.2 years) took part in this study. In the caseof one patient, the other eye was amblyopic and thereforecould not be tested as the visual acuity was very poor. Inanother patient, only one eye was tested as the patient wasvery tired and uncooperative. The clinical data of allpatients tested are listed in Tables 1—4. Most patients hadno other systemic problems but 5 patients suffered fromhypertension, ischaemic heart disease (IHD), hernia.

234 Ophthat. Physiot. Opt. 1997 17: No 3

Table 1. Clinical data of patients in Group 1

No.

1.

2.

3,

Table 2,

No.

1.

2.

3,

4,

Table 3.

No.

1.

2,

3.

4,

CJl

6.

7,

8,

'Age

50

50

52

Clinical

Age

60

70

50

52

Clinical

Age

64

54

58

56

52

68

67

70

Sex

F

F

F

data of

Sex

F

M

F

F

data of

Sex

M

F

M

F

F

F

F

F

Eye

RL

RL

RL

patients

Eye

RLRL

RL

R

patients

Eye

RLRL

RL

RL

R

RL

RL

RL

Duration

Syst.

2 yr

2 yr

2 yr

in Group 2

Ocu.

2 yr

2 yr

2 mth

Duration

Syst.

2 yr

6 yr

2 yr

1 yr

in Group 3

Ocu.

2 yr

6 yr

2 yr

1 yr

Duration

Syst.

25 yr

21 yr

6 mth

6 yr

2 yr7 yr

29 yr

1 yr

Oou.

22 yr

21 yr

6 mth

3 yr

2 yr

6,5 yr

3 yr

1 yr

VA

6/96/96/66/66/96/9

VA

6/96/96/66/66/96/96/6

VA

6/66/66/66/66/66/66/66/66/9

6/66/66/96/96/96/6

Exoph.

(mm)

CD C

D

1010

2424

Exoph.

(mm)

20202222

1314

16

Exoph.

(mm)

282826252421

262521

2932

IO

N)

CO C

O

1413

top

22232223

CO C

O

IOP

1°

20162020141815

top

22221417

CD O

CM CM

202024

CO toCN CM

2016

CO CO

CM CM

Up

303424281821

Up

302627281428

18

Up

24302128

4035302638

404034303232

Fields

NormalNormalNormalNormalNormalNormal

Fields

NormalNormalNormalNormalNormalNormalNormal

Fields

NormalNormalDefectDefectNormalNormal

DefectDefectDefectNormalNormal

DefectDefectNormalNormal

FMWOHue

error score

112116102110

139113

FMWOHue

error score

120140107104

9088

110

FMWOHue

error score

13872

50240412884

268286173

13987

280108180208

arthritis, supraventricular tachycardia or retinal veinthrombosis. All patients had visual acuity 6/9 or better.Thirteen patients had normal visual field plots. Only 2patients had defective visual fields. Both these patients hada relative defect in the inferior field when tested using theOctopus (500E), There was one patient who had no visualfield testing. Average pupil diameters of these patients were

2 mm. All patients were examined and were under themanagement of a Senior Registrar at the MREH.

1 Hz and 25 Hz spectral sensitivities of 10 normal subjectswho had no ocular or systemic disease were also tested ina similar way as the patients in this study. The age rangeof these normals were between 50-80 years (mean age =58 years, SD ^ 9.9). Their visual acuity was 6/6 or better.

spectral sensitivity in patients with dysthyroid eye disease: Sharanjeet-Kaur et al. 235

Table 4. Mean values (and SD) of the clinical data of all dysthyroid eye disease patients

Group

1

2

3

Age

50.7{1 .2 }

58,0(9.1)61.1(7.0)

Duration

Systemic

2.0(0)2.8

(2.3)

11.4(11.7)

Ocular

1.4(1.1)2.8(2.3)7.4(8.9)

Exoph.(mm)

17.8(6.1)18.1(3.8)24.5(5,3)

IOP1°

21.0(2.3)17.6(2.6)21.4(3.9)

Up

25.8(6.0)24.4(6.0)32.0(5,9)

FMWO hueerror score

115(12.5)

108(17.8)204

(124,2)

Testing procedure

Spectral sensitivity measurements were determined to a I °spectral light presented within a dark annulus of widthlOmlnoiarcupona 10° 1000 td white background (F/gMre2), The dark annulus of width 10 min of arc has been shownto segregate the chromatic and achromatic mechanismseffectively by depressing the sensitivity of the achromaticmechanism (Sharanjeet-Kaur, 1991). The wavelength ofthe test target was varied from 402—652 nm in approxi-mately 25 nm steps. The subject was optimally correctedfor distant vision and viewing was monocular.

The subject adapted for I min to the white backgroundprior to testing. Detection thresholds were determined to a1 Hz test stimulus by a method of adjustment. The subjectwas presented with an above threshold intensity test spot.The subject then reduced the intensity until the test spot justdisappeared. Flicker thresholds were determined to a 25 Hztest stimulus by a method of adjustment. The intensity ofthe flickering test spot was presented above threshold tothe subject and the subject reduced the intensity until theflickering just stopped. Short wavelength stimuli were

test

background

Figure 2. Schematic diagram of the test stimulus presentedto the subject on a targe background with a dark annulus(represented by z). The width of the dark annulus usedwas 10 min of arc.

presented first, beginning at 402 nm and then at approxi-mately 25 nm intervals to 652 nm. The sequence of 11stimulus presentations was repeated twice and the averagetaken.

FM l(X)-Hue test was also carried out monocularly oneach eye at a distance of 50 cm under a Macbeth DaylightEquivalent lamp of colour temperature 6700 K, illuminantC with an overall light level of 300 lux. This test was doneto compare it with spectral sensitivity measurements inorder to see if it was better, worse or just as effective asspectral sensitivity testing in detecting any colour visionabnormality.

Results

Spectral sensitivity recordings of dysthyroid eye diseasepatients showed that they could be divided into 3 maingroups:

Group 1: reduced 1 Hz curve but normal 25 Hz eurve.Group 2: normal 1 Hz curve but reduced 25 Hz eurve,Group 3: reduced 1 Hz and 25 Hz curves.

Figures J, 4 and 5 show the 1 Hz and 25 Hz spectralsensitivity curves for all the dysthyroid eye disease patientswho can be classified into Group 1, Group 2 and Group 3respectively. The continuous 1 Hz and 25 Hz spectralsensitivity curves with no symbols are the averages ofspectral sensitivities of 10 normal subjects. The dashedlines are ± one standard deviation. As long as the mean logsensitivity recordings of the 11 wavelengths of the dys-thyroid eye disease patients were within two standarddeviations of the normal data, it was considered normal.But if the mean log sensitivity of the 11 wavelengths wasdepressed by more than two standard deviations, then itwas considered to be abnormal. This criteria was used forall patient data analysis.

Tables 7, 2 and 3 show the clinical data of patients inGroups 1, 2 and 3 respectively and Table 4 shows the meanvalues of the data collected for all the three groups ofpatients.

Patients in Group I have reduced 1 Hz spectral sensitivityonly whereas patients in Group 2 have reduced 25 Hz

236 Ophthal. Physiol. Opt. 1997 17: No 3

Group I : reduced I Hz and normal 25 Hz

400 500 600 700 SOOWavelength (nm)

Figure 3. Graph of 1 Hz and 25 Hz spectral sensitivity forpatients in Group 1 (denoted by circle symbols). Thecontinuous lines without symbols are spectral sensitivitycurves of 1 Hz and 25 Hz presentation obtained from age-matched normals. The dashed line is ± 1 SD. The largeststandard errors are shown by error bars.

Group 2 : normal I Hz and reduced 25 Hz

400 500 600 700 800

Wavelength (nm)

Figure 4. Graph of 1 Hz and 25 Hz spectral sensitivity forpatients in Group 2 (denoted by circle symbols). Thecontinuous lines without symbols are spectra! sensitivitycurves of 1 Hz and 25 Hz presentation obtained from age-matched normals. The dashed line is ± 1 SD. The largeststandard errors are shown by error bars.

spectral sensitivity only. The mean systemie and oeulardurations of the disease of patients in these two groups isless than 3 years. These patients also have no proptosis.The mean exophthalmomeler readings are less than 22 mm.The mean intraocular pressures in primary gaze are also nothigh. i.e. less than 24mniHg, However, in upgaze, theintraocular pressures are slightly higher, i.e. greater than24mmHg(referto Table 4). TUe FM 100-Hue error scoresare also higher than the normal age matched value of 94.4(Pinckers. 1980). There was no particular confusion axis inthe FM 100 Hue plots. Patients in Groups 1 and 2 havenormal visual field plots.

Group 3 : reduced I Hz and 25 Hz

4r-

400 500 600Wavelength (nm)

700 800

Figure 5. Graph of 1 Hz and 25 Hz spectral sensitivity forpatients in Group 3 (denoted by circle symbols). Thecontinuous lines without symbols are spectral sensitivitycurves of 1 Hz and 25 Hz presentation obtained from age-matched normals. The dashed line is ± 1 SD, The largeststandard errors are shown by error bars.

Patients in Group 3 have reduced both 1 Hz and 25 Hzspectral sensitivities. These patients have a longer meansystemic and ocular duration of the disease (i.e. 11.4 yearsand 7.4 years respectively). There is also marked proptosisin this group of patients (mean exophthalmometer reading =24.5 mm). The intraocular pressure of patients in Group 3is only high on upgaze, i.e. mean = 32mmHg. The FMHue-100 error scores of these patients are also highercompared to their age-matched normals. The FM 100-Hueplots showed no particular confusion axis. Four out of eightpatients tested in Group 3 had defective visual fields.

Discussion

It is evident from the study that patients with dysthyroid eyedisease have reduced spectral sensitivities, either 1 Hz,25 Hz or both. Comparing between patients in Groups 1and 2 and patients in Group 3, it can be seen that patientsin Group 3 have longer systemic and ocular durations of thedisease. The FM 100-Hue error scores are also higher forpatients in Group 3 as compared to those in Groups 1 and2, Patients in Group 3 have statistically significant greaterexophthalmometer readings as compared to patients inGroup I {P < 0.01) or Group 2 ( / ' < 0.01). There are alsostatistically significant higher intraocular pressure read-ings in primary gaze (P < 0.02)) and upgaze (P < 0.01)between patients in Group 2 and 3. It appears that thegreater the loss in spectral sensitivity {as in patients inGroup 3), the larger the proptosis and the greater theintraocular pressure on upgaze.

Within the optic nerve, there are nerve fibres with largeand fast conducting axons and nerve fibres with small andslow conducting axons. The large fibres determine 25 Hzachromatic sensitivity whereas the small fibres determine

Spectral sensitivity in patients with dysthyroid eye disease: Sharanjeet-Kaur et al, 237

1 Hz chromatic sensitivity. This has been shown frompsychophysical experiments (King-Smith and Carden, 1976;Schwartz, 1992, 1993, 1995), physiological experiments(DeMonasterio and Gouras, 1975; Dreher el ai, 1976;Schiller and Malpeli, 1978) and lesion studies (Meriganetal.. 1978; Schiller e/a/,. 1990),

The data reveals that both niagno and parvo fibres can belost in dysthyroid eye disease. However, it is difficult todeduce which type of damage occurs first.

We believe that patients in Group 1 have some form ofdamage to the small fibres. These patients have a reduced1 Hz spectral sensitivity only. It is possible that in initialstages, the small fibres relaying chromatic information areaffected, probably due to reduced axoplasmic flow or re-duced blood supply to the smaller fibres (possible ischaemicmechanism).

It is important to note that chromatic opponency, asrevealed by the presence of the notch at 575 nm, is preservedin these cases. Sensitivity in this region is mediated byganglion cells carrying red-green information. Lee el al.(1987) have shown that the human spectral sensitivity func-tion matches the characteristics of chromatic cells in thelateral geniculate nucleus (LGN). In normals, there are manymore red-green than yellow-blue neurons. If both wereequally susceptible to the disease process then a greater lossof sensitivity in the blue-yellow region would be expected ashas been reported in glaucoma. It may be then, that in theearly stages of dysthyroid eye disease, red-green fibres areparticularly prone to damage. Whatever the mechanism,this would not lead to change in shape of the I Hz function;selective damage to fibres leads to reduced sensitivity in aparticular region of the spectrum. A change in the shape ofthe 1 Hz function would imply that though damaged, thefibres were capable of carrying luminance information, butnot colour information. This would be extremely unlikely.

As the dysthyroid eye disease progresses, it is possible thatthe large fibres are damaged. This could be caused indirectlyby enlargement of the extraocular muscles and proliferationof orbital fat and connective tissue resulting in an increasein the size of the intraorbital contents and therefore theintraocular pressure. It is possible that there is direct com-pression of the large fibres probably due to the indirecteffect of raised ocular pressure (possible mechanical mech-anism). It has been shown that in glaucoma, where there israised intraocular pressure, pressure on the optic nerve headresults in preferential damage to the large fibres (Gasser andErlanger, 1929; Quigley etal.. 1986; Glovinsky eta/.. 1991).

Spectral sensitivity plots of patients with dysthyroid eyedisease do show some damage to nerve fibres signallingchromatic and achromatic responses, even in patients witha short duration of the disease. In the more advancedcondition, as seen in patients in Group 3, it is possiblethat both the small and large fibres are damaged therebyreducing I Hz and 25 Hz spectral sensitivities.

Spectral sensitivity testing is a sensitive technique to

detect nerve fibre damage. It can be seen from the data(refer to Tables 1-3) that visual field plots are normal inpatients in Groups I and 2. In actual fact these patients havealready suffered some form of nerve fibre damage asrevealed by spectra! sensitivity testing. Even visual fieldplots of patients in Group 3 are normal in 50% of the cases.According to spectral sensitivity testing, it is shown that thenerve fibre damage is quite severe.

The severity of the nerve fibre damage is also revealedusing the FM 100-Hue testing. The error scores in all threegroups of patients are higher than the age-matched normals.The error scores of patients in Group 3 arc also higher thanthose in Groups 1 and 2. However, the FM 100-Hue plotsdo not reveal any confusion axis. Also, the plots do notindicate the type of nerve fibre damage, unlike the spectralsensitivity plots.

Acknuwledgement

We thank Dr O'Donoghue of the Manchester Royal EyeHospital for his help and support in this project.

References

Bedford. R. R. and Wyszecki, G. (1957). Axial chromaticaberration of the human eye. / Opt. Soc. Am. 47{6), 564-565.

Brown. J., Coburn, J, W,. Wigod, R. A. ei al- (1963),Adrenal steroids therapy of severe intiltralivcophthaimopathy ol'Graves disease. Am. J. Med. 34, 786—795,

DeMonasterio. F, M, and Gouras, P, (1975). Functionalproperties of ganglion cells of the rhesus monkey. J.Physiol. 251, 167-1%.

Dreher. B., Fukada. Y. and RtxJieck. R, W. (1976). Identi-fication, classification and anatomical segregation of cells,with X-like and Y-like properties in the lateral geniculatenucleus of Old World primates. / Physiol. 258, 433-452.

Gasser, H, S, and Erlanger. J. (1929). The role of fibre sizein the establishment of a nerve block by the pressure ofcocaine. Am. J. Physiol. 88, 581-591,

Glovinsky. Y.. Quigley. H. A, and Dunkelberger. G, R,(1991), Retinal ganglion cell loss is size dependent inexperimental glaucoma. Invest. Ophthalmol. and Visual Sci.32(3). 484-491.

Gorman, C. A., Desanto. L, W.. MacCarty. C. S, and Riley,F. C, (1974). Optic neuropathy of Graves disease.Treatment by transantral or transfrontal orbital compression.N. Engl. J. Med. 290, 70-75.

Kanski, J. J. and McAllister. J. A, (1989). Glaucoma. Acolour manual of dia^nosi.s and ireuimenl. Butterworths,London, UK.

King-Smith. P. E. and Carden, D, (1976), Luminance andopponent colour contributions to visual detection andadaptation and to temporal and spatial integration, J. Opt.Sac. Am. 66, 709-717.

Lee, B. B.. Valberg, A., Tigwell, D. A. and Tryti. S, (1987), Anaccountof responses of spectrally opponent neurons in macaquelateral geniculate nucleus to successive contrast, Prtx;eedingsof the Royal Society of Umdon. Series B. 230. 293-314.

Lewis, A. L., Katz. M. and Oehrlein, C. (1982). A modifiedachromatising lens. Am. J. Optom. and Physiol. Optics59(11). 909-911,

238 Ophthal. Physiol. Opt. 1997 17: No 3

Limberg. J, V. and Anderson, R, L. (1981), Transorbitaldecompression. Indications and results. Arch. Ophthalmol.99, 113-119.

Merigan, W. H., Katz, L. M. and Maunsell, J. H. R, (1989).ln\e.st. Ophthalmol. Visual Sci. 30(Suppl,). 53.

Panzo, G. J, and Tomsak, R, L, (1983), A retrospective reviewof 26 cases of dysthyroid optic neuropathy. Am. J.Ophthalmol. 96, 190-194.

Paulus. W. and Plendl, H, (1991). Colour vision defects inoptic nerve compression tested in the D and H colour rule.In: Colour vision deficiencies X. (eds. Drum, B., Moreland.J, D. and Sena, A.), Kluwer Academic Publishers. Dordrecht,

Pinckers. A. (1980). Colour vision and age. Ophthalmologica(Basel) 181, 23-30,

Quigley. H. A., Dunkelberger, G, R. and Sanchez, R. M.(1986). Chronic experimental glaucoma causes selectivelygreater loss of larger optic nerve fibres. ARVO abstracts.Invest. Ophthalmol. Visual Sci. 27(Suppl.), 42.

Ravin, J. G.. Sisson. J. C, and Knapp. W, T, (1975), Orbitalradiation for the ocular changes in Graves' disease. Am. J.Ophthalmol. 79, 285-288.

Riley. F. C, (1972). Surgical management of ophthalmology inGraves disease. Mayo Clin. Proc. 47(12) 986-988.

Schiller. P. H. and Malpeli, J, G, (1978). Functionalspecificity of lateral geniculate nucleus laminae of the rhesusmonkey. / Neurophysiol. 41, 788-797,

Schiller, P. H.. Logothesis, N. K. and Charles. E. R. (1990).Functions of the colour-opponent and broad-band channels ofthe visual system. Nature 343, 68-70,

Schwartz, S. H. (1992), Reaction time distributions and theirrelationship to the transient/sustained nature of the neuraldischarge. Vision Res. 32(11), 2087-2092.

Schwartz. S, H. (1993), Colour and flicker thresholds for highfrequency increments. Ophthalmic. Phvsiol. Opt. 13(3).299-302.

Schwartz. S, H, (1995), Spectral sensitivity of dichromats: roleof postreceptoral processes. Vi.sion Res. 34(22), 2983-2990.

Sergott, R. C. and Glaser, J. S, (1981). Graves ophthai-mopathy. A clinical and immunologic review, Surv.Ophthalmol. 26(1) 1-21,

Sharanjeet-Kaur (1991). Chromatic and Achromatic Mechanismsin Visual Processing. PhD Thesis. University of Manchester,UK,

Trobe, J, D,. Glaser. J, S, and Laflamme. P, (1978),Dysthyroid optic neuropathy. Clinical profile and rationalefor management. Arch. Ophthalmol. 96, 1199-1209.

Westheimer, G. (1966). The Maxweilian view. Vision Res. 6,669-682,

Zisman, F,, Carden, D,. Baxendale, K. and King-Smith. P, E.(1977). Techniques, instruments, cases. Am. J. Optom. andPhysiol. Optics 54(1). 52-55,