spheroidal degeneration of cornea and conjunctiva · brit. 7. ophthal. ( 976) 60, 473 spheroidal...

Brit. 7. Ophthal. ( 976) 60, 473

Spheroidal degeneration of cornea and conjunctiva

A. GARNER, F. T. FRAUNFELDER,* T. C. BARRAS, AND E. N. HINZPETERtFrom the Department of Pathology, Institute of Ophthalmology,UJniversity of London, and the External Disease Unit, Moorfields Eve Hospital, London

Small golden to yellow spherules in the cornea andconjunctiva have now been seen in people in everycontinent. There are, however, variations in thesize, distribution, and number of the deposits,governed probably by environmental and nutri-tional factors, the duration of the condition, andthe presence of other corneal lesions. Spheroidaldeposits occurring in a band-shaped pattern acrossthe central cornea are most common in regions inwhich the eyes are exposed to climatic extremes andthe traumatic effects of wind-blown sand or ice(Bietti, Guerra, and Ferraris de Gaspare, 1955;Freedman, I965; Freedman, I973b; Rodger,I973; Young and Finlay, 1975). In geographicalareas subject to much sunlight but spared thetraumatic effects of wind-blown sand or ice it iscommon to see other concomitant corneal orconjunctival lesions, particularly pingueculae (Klint-worth, 1972; Fraunfelder and Hanna, I973), whilein countries blessed with little actinic radiationthis may almost be the rule (Garner, 1970, I972).The prevalence of spheroidal degeneration intemperate climes characterized by low levels ofsunlight has, however, not been reported, and thepurpose of this paper is to document its incidencein the cornea and conjunctiva as seen in an Englisheye hospital and to comment on its associationwith pinguecula.

Patients and methodsFrom January to July I975 almost iooo random patientsseen in the outpatient clinic at Moorfields Eye Hospital,London, were examined using a slit lamp for the pres-ence of corneal or conjunctival droplets (Fig. i).Patients who had not spent at least three-quarters oftheir lives in Britain were excluded from the study.Photographs were taken using a macro camera (BrownI970) and a Zeiss photoslit lamp.Thirty-two eye bank eyes and five corneas known to

*Visiting Professor, Institute of Ophthalmology. Present address:Department of Ophthalmology, University of Arkansas MedicalCenter, 4301 West Markham, Little Rock, Arkansas 72201

tVisiting Worker, Institute of Ophthalmology. Present address:2 Hamburg 20, Eisenlohrsweg 4, West Germany

Address for reprints: Dr A. Garner, Institute of Ophthalmology,Judd Street, London WCIT-I oQS

contain spheroidal deposits were examined histologi-cally, the tissue being fixed in formol saline and embed-ded in paraffin wax. Staining procedures includedhaematoxylin and eosin, Verhoeff-van Gieson, orcein,and Gomori's aldehyde fuchsin on untreated sectionsand sections predigested for six hours with pancreaticelastase (Fullmer, I960).

Results

The prevalence of conjunctival and corneal spheroi-dal droplets at various ages is shown in Fig. 2. Theoverall prevalence was 6-5 per cent with a predomi-nance of males in patients over 50 years old in theratio of 7 men to 3 women. Droplets were four timesmore common in those who spent much of theirtime out of doors than in patients who lived pre-dominantly indoors. Analysis of the depositsaccording to the criteria introduced by Fraunfelderand Hanna (I973) showed that each of the 62affected patients had conjunctival droplets whileI2 also had corneal droplets. There was no sign ofprevious lesions in any of these patients.

Histological study was directed primarily to theconnexion between spheroidal droplets and thealtered connective tissue fibres seen in the fre-quently associated pingueculae (Fig. 3). Nine ofthe eye bank eyes had conjunctival pingueculaeand three of these also included spheroidal deposits.The coarse curled fibres of pingueculae stainedwith each of the elastic stains used, and whereaspretreatment with elastase had minimal effect onthe subsequent response to Verhoeff's stain orGomori's aldehyde fuchsin technique it usuallyprevented staining with orcein (Figs 4 to 6). Thespheroidal droplets, on the other hand, failed toreact with either orcein or aldehyde fuchsin irres-pective of predigestion with elastase. Their positiveresponse to Verhoeff's stain was also unaffected byelastase (Figs 7 to 9). The results are collated inTable I.

DiscussionEPIDEMIOLOGY

A comparison of the English data with the resultsof an American study in the State of Arkansas

on May 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.60.6.473 on 1 June 1976. D

ownloaded from

http://bjo.bmj.com/

474 British 7ournal of Ophthalmology

FIG. I Clinical photograph showing multiple droplets at corneal periphery

conducted according to the same criteria (Fraun-felder and Hanna, I973) shows a considerabledifference at all ages in the prevalence of spheroidaldegeneration (Fig. io). Of the environmentalfactors that were considered in the aetiology of thedeposits only the amounts of actinic radiation inArkansas and Britain are significantly different.This, together with the finding that a majority ofthe English patients with droplets spent much oftheir lives out of doors, supports the suggestionthat exposure to sunlight is of pathogenic import-ance. Other geographical areas where the preva-lence of spheroidal degeneration is high also have,among other things, much sunlight (Table II).Possibly the prevalence of the deposits in the generalpopulation in England is less than would seemfrom our data since our study was undertaken onoutpatients to an eye hospital, who were thereforea selected group.

PATHOLOGY

Several workers have commented on the con-comitance of spheroidal degeneration in the corneaand pinguecula (Klintworth, I1972; Fraunfelderand Hanna, 1973) and their similar staining proper-

ties (Brownstein, Rodrigues, Fine, and Albert,I973; Rodrigues, Laibson, and Weinreb, 1975).Since apparently identical deposits have longbeen known to occur in the conjunctival tissueoverlying pingueculae (Fuchs, I89I; Parsons,I904; Hogan and Zimmerman, I962; Klintworth,1972) it has been proposed that both corneal and

30 28

* Male

U Female

4uL20

Nopaiet 3107 0 64 6IB 66 34

cr~~~~~~~~~3 12a

a.. 5

0 AE- 0 -0

Aqe 0-30 31-40 41-SO 51-60 61-70 71 +NoSpatients 48 130 72 70 464w4 9608 86 66 53 44

Spheroidal 2 0 4 4 0 2 10 0 10 6 i5degenerationFIG. 2 Prevalence of spheroidal droplets in conjunctivaandlor cornea in British subjects related to age and sex



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

Spheroidal degeneration 475

FIG. 3 Section showing hyaline droplets immediately beneath epithelium of corneal periphery associated with coarsedegenerate fibres of a bulbar conjunctival pinguecula. Verhoeff-van Gieson. x IIi

,- 2- V Nw#qrto:/

C".r"

(40) (4b)FIG. 4 Pinguecula. (a) Pseudoelastic fibres stain intensely with Verhoeff's haematoxylin. (b) Pretreatment with elastasefor 6 hours partially inhibits staining reaction but coarse convoluted fibres continue to react. Verhoeff-van Gieson. x 195

conjunctival droplets are the result of degenerativechanges in the stromal collagen (Klintworth,1972; Brownstein and others, 1973; Christensen,1973; Rodrigues and others, '975). This concepthas much in its favour. Our finding, however, thatdespite a common propensity to stain by Verhoeff'sprocedure for elastic the ability of pingueculae toreact with other elastic stains such as orcein and

aldehyde-fuchsin is not shared by the spheroidaldeposits, does suggest that the transition fromdegenerative collagen to spheroidal droplets isdue to more than a simple process of fibrillar disin-tegration. Adequate explanation of these stainingdifferences requires a better understanding of theelastotic change seen in pinguecula formation andmore infornation concerning the chemistry of the

4p-mATF ..-t

S.



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

476 British Journal of Ophthalmology

04

* N

FIG. 5 Pinguecula. Abnormal fibres stain blue withaldehyde-fuchsin and, apart from some less coarse fibres,are essentially unaffected by elastase. Gomori'saldehyde-fuchsin. x 195

staining reactions, but they do point to a differencein composition between the deposits and elastoticcollagen. The earlier finding of tryptophan andappreciable amounts of tyrosine and sulphur-containing amino-acids in the deposits (Garner,1970; Klintworth, 1972; Brownstein and others,1973; Garner, Morgan, and Tripathi, I973)underlines this disparity, since these amino-acidsare either absent or minimal in elastic tissue andcollagen. The electron microscopy studies ofHanna and Fraunfelder (I972) may be relevant inthis context, since they concluded that the spheroi-dal droplets represent the condensation of agranular material secreted by abnormal fibrocytesand deposited on collagen fibrils. Hogan andAlvarado (I967) have also commented on thepresence of a non-collagenous protein in thespheroidal deposits. The addition of a second

.' "

.& _ _

. r 1 v _:

(6a) (6b)FIG. 6 Pinguecula. Positive staining with orcein largely prevented by predigestion with elastase. (a) Without elastase;(b) with elastase. Orcein. x 195

_wE----,*l _==w9w _ _ .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~------

(701) (7b)FIG. 7 Spheroidal droplets in cornea stain intensely with Verhoeff's haematoxylin and are unaffected by pretreatmentwith elastase. (a) Before elastase; (b) after elastase. Verhoeff-van Gieson. x 195

oil,io mbw.



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

Spheroidal degeneration 477

protein, from whatever source, to the alteredcollagen fibres of a pinguecula (or to altered cornealstroma) might account for the somewhat differentstaining properties.With regard to the staining properties of the

pingueculae themselves, it is to be noted that weobserved variable but generally marginal loss ofability to react with Verhoeff's haematoxylin andaldehyde-fuchsin as a result of predigestion withelastase and complete inhibition of orcein stainingin such preparations. This effect of elastase is atslight variance with the usual statement that thefibres of a pinguecula are resistant to this type ofenzymatic activity (Cogan, Kuwabara, and Howard,1959) and is not easily explained, although it is of

FIG. 8 In contrast to fibres of pingueculae, spheroidaldroplets do not react with Gomori's aldehyde-fuchsin.X 130

a

FIG. 9 Section of spheroidal droplets in cornea showingabsence of staining with orcein. x 195

interest that Lever (I967) states that elastoticfibres in the skin are digested.

NOMENCLATURE

Increased interest on the part of a number ofinvestigators coupled with a persistent lack ofcrucial data has led to a profusion of names for thehyaline deposits. The term keratinoid, introducedby Garner (1970, 1972), has not found generalacceptance and, in the light of subsequent evidencelinking the deposits with elastotic degeneration ofcollagen, is probably best abandoned. Some recentterms such as proteinaceous degeneration (Chris-tensen, I973) are too vague, whereas others,such as actinic (Klintworth, 1972) or climatic(Freedman, I973a) keratopathy, are not only

Table I Staining reactions for elastic tissue in patients with pingueculae and/or corneal droplets

Spheroidal droplets

Before elastase After elastase VerhoeffCaseno. Aldehyde- Aldehyde- Before After

Verhoeff fuchsin Orcein Verhoeff fuchsin Orcein elastase elastase

+ +± ++

+ ±± + ± + ++ ± + + +

+ + + - +

Aldehyde- Orceinfuchsin

+

i+ + - _

Pingueculae

I

2

34 +567 +8 +9 +I0I I

12'314

71

'Ph



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

478 British Journal of Ophthalmiology

unproved but are also unduly restrictive sincethey do not account for the corneal deposits secon-dary to other corneal disease. Until the natureof the deposits is known and their aetiology sub-stantiated it may be better to use a purely descrip-tive term such as spheroidal degeneration ordroplet keratopathy.

SummaryA study of almost iooo outpatients at a Londoneye hospital showed the presence of asymptomaticyellowish, spheroidal deposits in the peripheralcornea or conjunctiva, or both, in about 6 per cent,with a preponderance of males and older subjects

75

CI

. 50

0

cr0

a.-,,- 2 5-

Age 0-3CNo. patients 278 128!Spheroidal 2

deqe neration

IGreat BritainUnited States_ ~~~~~~~~~~~~~~53

38 40

22 2

) 31-40 41-50 51-6O 61-70 71 +5 142 58 90 361 204 385 152 245 98 2650 8 4 2 7X 12 147 16 97 22 141

FIG. 10 Comparison of prevalence of spheroidaldroplets in conjunctiva and/or cornea in Great Britainand United States of America at various ages

Table II Reported prevalence of spheroidal dropletsin various parts of the world (not age- or sex-matched)

Geographicallocation

Prevalenceof spheroidaldroplets ingeneralpopulation Reference(percentage)

Dahlak Island, 57Ethiopia

Nama, South-West 3 1Africa

Arkansas, USA I8

Labrador, Canada iILabrador, Canada I4

London, England 6

Rodger, 1973

Freedman, J., 1973b

Fraunfelder andHanna, I973

Freedman, A., I965Young and Finlay,

I975Present study

among those affected. This prevalence is less thanis observed in people in countries exposed to higherlevels of sunlight.

Histological study of the deposits showed sometinctorial similarities with the pseudoelastic fibresof pingueculae, with which they were sometimesassociated, but also significant differences suggest-ing that the spheroidal deposits might be a com-posite of degenerate collagen and a second non-collagenous protein.

Until a more precise terminology is feasible wesuggest that a purely descriptive name such asspheroidal degeneration or droplet keratopathyshould be used to describe this entity.

ReferencesBIETTI, G. B., GUERRA, P., and FERRARIS DE GASPARE, P. F. (I955) Bull. Soc. franc. Ophtal., 68, 10IBROWN, N. A. (1970) Brit. J3. Ophthal., 54, 697BROWNSTEIN, S., RODRIGUES, M. M., FINE, B. S., and ALBERT, E. N. (1973) Amer. J. Ophthal., 75, 799CHRISTENSEN, G. R. (I973) Arch. Ophthal. (Chic.), 89, 30COGAN, D. G., KUWABARA, T., and HOWARD, J. (1959) Ibid., 6i, 388FREEDMAN, A. (I965) Ibid., 74, I98

(1973a) Ibid., 89, I93FREEDMAN, J. (Ig73b) Brit. J3. Ophthal., 57, 688FRAUNFELDER, F. T., and HANNA, C. (1973) Amer. Y. Ophthal., 76, 41FUCHS, E. (I891) v. Graefes Arch. Ophthal., 37, I43FULLMER, H. M. (I960) Y. Histochem. Cytochem., 8, 290GARNER, A. (1970) Brit. J. Ophthal., 54, 769

(1972) Eye, Ear, Nose Thr. Monthly, 51, 299, MORGAN, G., and TRIPATHI, R. C. (i973) Arch. Ophthal. (Chic.), 89, I98

HANNA, C., and FRAUNFELDER, F. T. (1972) Amer. J. Ophthal., 74, 829HOGAN, M. J., and ALVARADO, J. (1967) Arch. Ophthal. (Chic.), 78, 174

, and ZIMMERMAN, L. E. (i962) 'Ophthalmic Pathology', 2nd ed., p. 253. Saunders, PhiladelphiaKLINTWORTH, G. K. (i172) Amer. J. Path., 67, 327LEVER, W. F. (i967) 'Histopathology of the Skin', 4th ed., p. 266. Lipincott, PhiladelphiaPARSONS, J. IH. (1904) 'The Pathology of the Eye', vol. I, p. 97. Hodder & Stoughton, LondonRODGER, F. C. (I973) Brit. J. Ophthal., 57, 657RODRIGUES, M. M., LAIBSON, P. R., and WEINREB, S. (1975) Arch. Ophthal. (Chic.), 93, IIIYOUNG, J. D. H., and FINLAY, R. D. (1975) Amer. J. Ophthal., 79, 129



j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

spheroidal degeneration of cornea and conjunctiva · brit. 7. ophthal. ( 976) 60, 473 spheroidal...

Documents