PostScript - British Journal of Ophthalmologytive best corrected visual acuity (BCVA) was 3/200...

PostScript..............................................................................................

LETTERS

Laser induced chorioretinalvenous anastomosis in ischaemiccentral retinal vein occlusionLaser induced chorioretinal venous anas-tomosis (CRVA) has been advocated byMcAllister and Constable as a treatment fornon-ischaemic central retinal vein occlusion(CRVO).1 This technique potentially offers ameans of permanently bypassing the site ofobstruction to venous outflow, which isthought to occur in the region of the laminacribrosa. In ischaemic CRVO, the visualprognosis is usually much poorer, with devas-tating complications like neovascular glau-coma and progressive macular ischaemia.2 Inthis prospective study, we investigated thefeasibility of laser induced CRVA in eyes withischaemic CRVO, in view of the possibility ofavoiding or lessening these severe complica-tions.

Materials and methodsThe classification of ischaemic CRVO wasbased on the presence of 10 disc diameter ormore of capillary non-perfusion in the fundusfluorescence angiography (FFA), according tothe criteria in the CRVO study.2 Approval fromthe ethics committee and informed consentfrom patients were obtained. Inclusion andexclusion criteria are shown in Table 1. All thelaser treatment was performed by one of theauthors (AK) who had successfully treatedpatients with non-ischaemic CRVO with a

similar procedure. The site for attempts at thecreation of anastomosis was in the inferotem-poral and superonasal retina over a venoustributary of the retinal vein where it crossesover an underlying choroidal vein, at least 3disc diameters away from the optic disc.Argon or diode laser with 50 µm spot size of0.1–0.2 second’s duration and with a powerlevel of 1.5–2.5 W was focused over the edge ofthe chosen retinal vein. Increasing power wasused until there was haemorrhaging from thevein (Fig 1A). The bleeding was stopped bypressure on the eye with a contact lens.

ResultsSix eyes of six patients were included (Table2). All of them had posterior vitreous detach-ment. Median follow up was 21 months(range 5–31 months). The median preopera-tive best corrected visual acuity (BCVA) was3/200 (range, hand movement to 8/200). Themedian postoperative best corrected visualacuity (BCVA) was 2/200 (range, hand move-ment to 20/200). The median number ofattempted anastomosis sites per eye was four(range, two to four). Through repeated oph-thalmoscopic examination, FFA, and indocya-nine green angiography, no functional anasto-mosis was found. A small nodular fibrotic scarwas noted in each site (Fig 1B). No other sig-nificant laser related complication was found.One eye eventually developed rubeotic glau-coma.

CommentIn non-ischaemic CRVO, a successful CRVAwas created in 33–54% of eyes.1 3 Laser

photocoagulation treatment parametersdiffered, because the superiority of onecombination of parameters compared withanother had not been demonstrated.3 In ourstudy, it appears that argon or diode laserinduced CRVA was not feasible in ischaemicCRVO. We attribute this to the severeendothelial cell damage secondary to ischae-mia and venous thrombosis across the retinalcirculation. In a dog model without retinalvein occlusion, a successful laser inducedCRVA was shown to be lined by endothelialcells.4 Despite the failure to create functionalCRVA, we did not encounter any adversecomplication related to the laser treatment.The presence of posterior vitreous detach-ment in our patients might have lessened thechance of development of chorioretinovitrealneovascularisation. Successful CRVA inischaemic CRVO has been reported to beestablished through pars plana vitrectomywith direct surgical puncture or erbium:YAGlaser.5–7 This surgical approach may be abetter option to create CRVA in ischaemiceyes, especially when the posterior hyaloid isstill attached preoperatively.

Table 1 Inclusion and exclusion criteria of patients

Inclusion criteria:1 Confirmed presence of central retinal vein occlusion2 Central retinal vein occlusion <3 months’ duration3 Visual acuity </=20/2004 Intraocular pressure <30 mm Hg5 Ability to obtain good quality fundus photographs and angiograms6 Age >21 years old

Exclusion criteria:1 Intercurrent eye disease of study eye that is likely to affect visual acuity over study period2 Presence of any diabetic retinopathy in either eye3 New or old branch artery/vein occlusion in study eye4 Other retinal vascular disease in study eye5 Vitreous haemorrhage other than breakthrough in study eye6 Presence of neovascularisation of the study eye (iris, angle, retina, disc)7 Heparin/warfarin sodium cannot be discontinued for duration of study8 Impossible to differentiate between ischaemic and non-ischaemic central retinal vein occlusion

Table 2 Baseline and outcome characteristics of patients receiving laser treatment

PatientNo Sex/age

Interval of CRVOand laser (weeks)

Laserused

No of lasersessions

Total No of lasersites attempted

InitialBCVA

FinalBCVA

Duration of followup (months) Complication

1 M/79 5 Argon 2 4 8/200 HM 31 Neovascular glaucoma2 F/80 2 Diode 1 2 2/200 2/200 313 F/54 3 Argon 2 4 4/200 20/200 194 M/53 12 Diode 2 4 8/200 5/200 235 F/80 6 Argon 2 4 HM HM 76 F/78 10 Argon 1 2 HM 2/200 5

CRVO = central retinal vein occlusion; BCVA = best corrected visual acuity.

Figure 1 Patient 3. (A) Retinal andsubretinal haemorrhage around theattempted anastomotic site along theinferotemporal retinal vein shortly after thelaser treatment. (B) Seven months later, asmall fibrotic nodule was present withminimal amount of altered blood.

Br J Ophthalmol 2003;87:1043–1055 1043

www.bjophthalmol.com

on August 24, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.87.8.1044 on 24 July 2003. D

ownloaded from

http://bjo.bmj.com/

Financial and proprietary interest: nil.

Financial support: nil.

A K H KwokDepartment of Ophthalmology, Hong Kong

Sanatorium and Hospital, Hong Kong,People’s Republic of China

A K H Kwok, V Y W Lee, T Y Y LaiDepartment of Ophthalmology and Visual Sciences,

The Chinese University of Hong Kong,Hong Kong Eye Hospital, Hong Kong,

People’s Republic of China

C HonDepartment of Ophthalmology, Queen Mary

Hospital, Hong Kong, People’s Republic of China

Correspondence to: Dr Alvin K H Kwok,Department of Ophthalmology, Hong Kong

Sanatorium and Hospital, 2 Village Road, HappyValley, Hong Kong; [email protected]

Accepted for publication 1 December 2002

References1 McAllister IL, Constable IJ. Laser-induced

chorioretinal venous anastomosis fornon-ischemic central retinal vein occlusion:evaluation of the complications and their riskfactors. Am J Ophthalmol 1998;126:219–29.

2 Central Vein Occlusion Study Group.Baseline and early natural history report: thecentral vein occlusion study. Arch Ophthalmol1993;111:1087–95.

3 Fekrat S, Goldberg MF, Finkelstein D.Laser-induced chorioretinal venousanastomosis for nonischemic central or branchretinal vein occlusion. Arch Ophthalmol1998;116:43–52.

4 Vijayasekaran S, Yu DY, McAllister IL, et al.Optimal conditions required for the creation ofan iatrogenic chorioretinal venousanastomosis in the dog using argon greenlaser photocoagulation. Curr Eye Res1995;14:63–70.

5 Fekrat S, de Juan E. Chorioretinal venousanastomosis for central retinal vein occlusion:transvitreal venipuncture. Ophthalmic SurgLasers 1999;30:53–5.

6 Peyman GA, Kishore K, Conway MD.Surgical chorioretinal venous anastomosis forischemic central retinal vein occlusion.Ophthalmic Surg Lasers 1999;30:605–14.

7 Quiroz-Mercado H, Sanchez-Buenfil E,Guerrero-Naranjo JL, et al. Successful erbium:YAG laser-induced chorioretinal venousanastomosis for the management of ischemiccentral retinal vein occlusion. A report of twocases. Graefes Arch Clin Exp Ophthalmol2001;239:872–5.

Actinic granuloma of theconjunctivaActinic granuloma is a condition character-ised, histologically, by a preponderance ofgiant cells in close relation with damagedelastic fibres and the absence of necrobiosis,lipid, mucin, and palisading of thegranuloma.1 The term was coined in 1975 by

O’Brien who described similar histologicalfeatures in cutaneous lesions of patients withsun damaged skin.2 Actinic elastosis is thehallmark of pingueculae and has been notedin association with a granulomatous reactionin only one previous report.3

Over the past three decades, three cases ofactinic granuloma of the conjunctiva havebeen documented in the literature (Table1).4–6

The paucity of reports ensures that the con-dition is under-recognised both clinically andpathologically. We describe a further case withthe novel association of scleral thinning, andfurther review the literature with reference topathogenesis, disease associations, treatment,and significance for both ophthalmologistsand pathologists.

Case historyA 67 year old white woman presented to theophthalmology department with a 3 weekhistory of a painless, red right eye. She had noprevious ophthalmic problems. Significantpast medical history included treated pulmo-nary tuberculosis and a lumpectomy forbreast carcinoma 10 years previously. She hadno significant family history and otherwisewas generally well.

Ocular examination demonstrated a 3 × 3mm, raised, flesh coloured, vascularised lesionon the right nasal bulbar conjunctiva with anassociated corneal dellen (Fig 1). Furtherscleral changes were observed circumferen-tially around the limbus under an apparentlyhealthy conjunctiva (Fig 1, arrowhead).

Initial differential diagnosis was carcinomain situ of the conjunctiva or metastasis frombreast carcinoma, although the lesion was notclinically typical of either.

An excision biopsy was performed. Duringsurgery the underlying sclera was noted to bedegenerate with significant thinning.

Histology of the lesion demonstrated dys-plasia within the squamous epithelium andprominent solar elastosis with a granuloma-tous response to degenerative elastic fibres.There was a lymphoplasmacytic infiltratecharacteristic of an inflamed pinguecula with

granulomatous features suggestive of actinicgranuloma (Fig 2).

Investigations into the cause of the under-lying focal scleral atrophy included full bloodcount, erythrocyte sedimentation rate, serumVDRL, serum complement, anti-ro and anti-laantibodies, and rheumatoid factor which wereall within normal limits. A screening serumANCA was weakly positive (1:20) but anti-myeloperoxidase assays were negative.

Anterior segment ultrasonography wasnormal. Fluorescein angiography of the ante-rior segment demonstrated an evenly per-fused iris, but a filling defect clearly deline-ated the atrophy and thinning at the lesionsite. Our patient was subsequently treatedwith topical antibiotics and eye padding andresponded favourably over several weeks withprogressive epithelialisation over the excisiondefect (Fig 3). The scleral changes persistedafter resolution of the epithelial defect (Fig 3,arrowhead).

CommentO’ Brien, in his original description of actinicgranuloma, described the pathogenesis as aphenomenon of repair occurring in damagedconnective tissue.2 This concept was disputedby Ragaz and Ackerman7 who believed thatthe granulomatous inflammation was not aresponse to degenerative elastotic fibres but

Table 1 Review of previously published cases of actinic granuloma

Patient age(years) Sex Location

Size oflesion Clinical presentation Differential diagnosis

Case 1, Proia et al4 38 Female Temporal bulbar conjunctiva 2 mm 3/52 History of painlessred eye

Vascularised pinguecula

Case 2, Ferry et al5 30 Female Temporal bulbar conjunctiva 2 × 3 mm 6/52 History of red eye Pingueculitis, contact lens reaction,conjunctival naevus

Case 3, Steffen et al6 39 Female Conjunctiva, site unknown Not known Not known Pingueculitis, Bowen’s diseaseCase 4, Gallagher et al 67 Female Nasal bulbar conjunctiva 3 × 3 mm 3/52 History of painless

red eyeBowen’s disease, breastmetastasis

Figure 1 Presenting appearance of lesionscleral thinning shown by arrowhead. Figure 2 Histology of excised lesion.

Figure 3 Appearance post-incision.Persisting scleral thinning shown byarrowhead.

1044 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

that the lesions described by O’Brien repre-sented variants of granuloma annulare, a dis-order of skin and ocular adnexae.8 9 The exist-ence of conjunctival actinic granulomas inisolation distinguishes this condition fromgranuloma annulare and implies that granu-loma formation can occur in response to elas-totic material. Furthermore, actinic granulo-mas are histologically distinct with prominentelastotic degeneration of connective tissuefibres, giant cells, and inconspicuous palisad-ing of epithelioid histiocytes.

McGrae postulated that actinic granulomarepresented a cell mediated immune responseto weakly antigenic determinants on actinic-ally altered elastotic fibres with a predomi-nance of helper T cells in the lymphocyticinfiltrate.1

More recently the association of temporalarteritis and actinic granulomas of the skinhas been documented.10 It is hypothesisedthat actinic radiation selectively injures elastictissue in the skin and its superficial arteriesand this tissue may then become antigenic,with local, humoral, and systemic overtones.

It is reported that the serum of patientswith untreated giant cell arteritis contains asignificantly elevated level of an elastase inthe form of matrix metalloproteinase 9(MMP-9) and that this enzyme was found tobe abundant in the vicinity of damagedtemporal internal elastic laminae.11 Gillot etal12 observed that sera from 12 of 13 patientswith untreated giant cell arteritis containedhigh levels of elastase derived elastin peptidesand that the peptides were targeted by T lym-phocytes such as appear in the actual lesionsof actinic granuloma1 and giant cellarteritis.13 This mode of autoimmune reactioncomplies with the “danger” model of autoim-munity described by Matzinger and appraisedby Larkin.14

Our case presented with the novel associ-ation of an underlying focal scleral atrophy.Negative investigations for scleritis wouldsuggest that this feature may be an extensionof the autoimmune process representative ofactinic granuloma rather than an independ-ent idiopathic scleritis.

It is interesting to note that all documentedcases of actinic granuloma of the conjunctivahave occurred in females which would besupportive of an autoimmune pathogenesis.Clinically, the differential diagnosis ofconjunctival actinic granuloma includespingueculitis, Bowen’s disease, conjunctivalnaevus, granuloma annulare (pseudorheu-matoid nodule), and episcleral rheumatoidnodule.

Pathologically, the differential diagnosisincludes pingueculae, pingueculitis,infection—particularly fungal, parasitic, ormycobacterial—and foreign body reactions.However, there is no granulomatous reactionto the actinic elastosis in pingueculae. In fun-gal and parasitic lesions there is often aprominent eosinophilic infiltrate associatedwith the granulomas. Caseous necrosis is seenin mycobacterial infections. In difficult casesspecial stains may help. Polarised light micro-scopy rules out the presence of any birefrin-gent material.

Actinic granuloma of the conjunctivarepresents a distinct clinical, histopathologi-cal, and immunological entity. Its classicpresentation over a short period of a fewweeks and poor response to topical steroidtreatment should aid the ophthalmologist inrecognising this lesion. Of practical import-ance to the ophthalmic pathologist is recog-nition that the granulomatous inflammationmay be associated with elastotic degenera-

tion and does not necessarily imply the pres-ence of a foreign body, fungal, or mycobacte-rial infection.

M J Gallagher, F Roberts, S Osborne,C M Kirkness

Tennent Institute of Ophthalmology, GartnavelGeneral Hospital, Great Western Road, Glasgow

G12 0YN, UK

Correspondence to: Stuart A Osborne, TennentInstitute of Ophthalmology, Gartnavel General

Hospital, Great Western Road, GlasgowG12 0YN, UK; [email protected]


References1 McGrae JD Jr. Actinic granuloma: a clinical,

histopathologic, and immunocytochemicalstudy. Arch Dermatol 1986;122:43–7.

2 O’Brien JP. Actinic granuloma: an annularconnective tissue disorder affecting sun- andheat-damaged (elastotic) skin. Arch Dermatol1975;111:460–6.

3 Bernardino V, Olivar E, Mangubat L, et al.Solitary nodular conjunctivitis-the inflamedpinguecula, granulomatous reaction toelastotic fibres. Philippine J Ophthalmol1972;4:1.

4 Proia AD, Browning DJ, Klintworth GK.Actinic granuloma of the conjunctiva. Am JOphthalmol 1983;96:116–18.

5 Ferry AP, Kaltreider SA, Wyatt DB. Actinicgranuloma of the conjunctiva. ArchOphthalmol 1984;102:1200–2.

6 Steffen C. Actinic granuloma of theconjunctiva. Am J Dermatopathol1992;14:253–4

7 Ragaz A, Ackerman AB. Is actinic granulomaa specific condition? Am J Dermatopathol1979;1:43–50

8 Rao NA, Font RL. Pseudorheumatoid nodulesof the ocular adnexae. Am J Ophthalmol1975;79:471–8.

9 Ferry AP. Subcutaneous granuloma annulare(“pseudorheumatoid nodule”) of the eyebrow.J Pediatr Ophthalmol 1997;14:154–7.

10 O’Brien JP, Regan W. Actinically degenerateelastic tissue is the likely antigenic basis ofactinic granuloma of the skin and of temporalarteritis. J Am Acad Dermatol1999;40:214–22.

11 Sorbi D, French DL, Nuovo GJ. Elevatedlevels of 92-kd type IV collagenase (matrixmetalloproteinase 9) in giant cell arteritis.Arthritis Rheum 1996;39:1747–53.

12 Gillot J-M,Masy E, Davril M. Elastase derivedelastin peptides: putative autoimmune targetsin giant cell arteritis. J Rheumatol1997;24:677–82.

13 Andersson R. Immunological studies in giantcell arteritis. Ballieres Clin Rheumatol1991;5:405–12.

14 Larkin M. Polly Matzinger: immunology’sdangerous thinker. Lancet 1197;350:38.

Unilateral nasal hemianopsiasecondary to posteriorsubcapsular cataractVisual field defects respecting the verticalmidline are a common occurrence associatedwith focal neurological lesions. However, uni-lateral nasal hemianopsias are rare defects,documented to be associated with pituitaryadenomas, temporal optic nerve lesions, andsuprasellar aneurysms.1 Cataracts are knownto depress the overall sensitivity of the visualfield,2 3 but localised visual field defects due tocataract are extremely rare and, to our knowl-edge, only three other cases have beenreported in the literature.4–6 We report a case ofa right unilateral nasal hemianopsia resultingfrom central posterior subcapsular lens opac-ity.

Case reportA 51 year old woman treated for normal ten-sion glaucoma in her right eye for 2 yearsattended for a review of her glaucoma follow-ing a change of medication with the additionof bimonidine eye drops to dorzolamide eyedrops. At this 3 monthly review the patient

Figure 1 Humphrey 24-2 visual field showing (A) precataract visual field, (B) field withcataract and visual field defect, (C) visual field after cataract removal.

PostScript 1045




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

gave a 1 month history of a sudden onset ofmisty vision affecting her right nasal visualfield noticed while driving her car. There wereno other associated neurological symptoms.Just before this she had been diagnosed with“borderline” systemic hypertension. Therewere no other risk factors for a vascular event,although there is a positive family history—her father had had a cerebrovascular accident.

On examination, her visual acuity haddeteriorated from 6/6 to 6/24 in the right eye,remaining unchanged at 6/6 in the left sincethe previous visit. It had also been docu-mented that letters on the nasal side of theSnellen chart were not seen with the righteye. Confrontation visual field demonstrated anasal hemianopsia of the right eye. Her pupilswere equal with normal reactions to light andaccommodation. Dilated slit lamp biomicros-copy revealed marked central posterior sub-capsular lens opacity with very mild subcap-sular changes in the other eye, previouslydocumented as normal. Retinal examinationwas normal and the optic discs pathologicallycupped with inferior rim thinning; changesconsistent with glaucoma, although therewere no documented changes from the previ-ous visit 6 months earlier.

Further neurological and cardiovascularexamination, other than a blood pressure of170/70 mm Hg, were also unremarkable.

The right nasal hemianopsia respecting thevertical midline was confirmed on a clinicallyreliable Humphrey 24-2 plot with a change inmean deviation from −5.94 to −17.43 from thepreviously documented visual field (see Fig1). The visual field of the left eye was normal.Routine blood tests and chest x ray were nor-mal. A computed tomograph (CT) scan of thebrain, orbits and visual pathways was alsounremarkable. In the absence of a focalneurological lesion this woman subsequentlyunderwent an uncomplicated right phacoe-mulsification and intraocular lens replace-ment. A repeat red spot visual field revealedcomplete reversal of the previously docu-mented right nasal hemianopsia, and a resto-ration of the visual acuity to 6/6.

CommentMedia opacities are known to cause visualfield defects,7 the degree of which varies fromgeneralised depression of the visual field8 toapparent scotomatous areas. Localised parax-ial lens opacities causing defects mimickingneurological abnormalities are extremelyrare. These opacities necessitate a posteriorposition in the lens to produce a relativescotoma. An opacity in the media anteriorlyplaced produce generalised reduction in thevisual field. In the previous three reportedcases all the cataracts were posterior subcap-sular in nature.

Our case is unusual in that the cataract wasplaced centrally, not temporally as may beexpected with a nasal defect. Further, the his-tory suggested a sudden onset which necessi-tated neurological examination and investiga-tion.

We feel it would have been inappropriate toproceed to surgical intervention withoutprevious investigation, which should alwaysinclude a CT scan (more preferably, if readilyavailable, an magnetic resonance imaging(MRI) scan), of the optic nerves and visualpathways. This unusual visual field defectmay have been present for some time beforethe appearance of the cataract but, if subtle,may not have been evident. However, thepresence of the cataract may have decreasedthe sensitivity of the eye and made the visualdefect more prominent. Our concern with this

woman was that removing the cataract mayhave disguised a more sinister underlyingpathology—that is, a neurological defect, sowe decided on a red spot visual field toconfirm that such a defect was indeed notpresent.

Whereas in this case the right nasalhemianopsia was due to a posterior subcapsu-lar cataract, we believe that a neurologicalcause for the field defect should always besought, particularly with a history of suddenonset. Cataract extraction should be consid-ered only establishing the absence of thesame. Further, consider performing a postop-erative red spot visual field test to confirm theabsence of focal neurology in such eyes.

I Rahman, A Nambiar, A F SpencerManchester Royal Eye Hospital, Lister Centre,

Nelson Street, Manchester, UK

Correspondence to: Mr Imran Rahman, ManchesterRoyal Eye Hospital, Lister Centre, Nelson Street,

Manchester, UK; [email protected]


References1 Miller NR, Newman NJ. Topical diagnosis of

lesions in the visual sensory pathway. In:Miller NR, Newman NJ, eds. Walsh andHoyt’s clinical neuro-ophthalmolgy. 5th ed.Baltimore: Williams and Wilkins,1998;1:302–3.

2 Fine EM, Rubin GS. Effects of cataract andscotoma on visual acuity: a simulation study.Optom Vis Sci 1999;76:468–73.

3 Fine EM, Rubin GS. The effects of simulatedcataract on reading with normal vision andsimulated central scotoma. Vis Res1999;39:4274–85.

4 Karp CL, Fazio JR. Traumatic cataractpresenting with unilateral nasal hemianopsia.J Cataract Refract Surg 1999;25:1302–3.

5 Phillips CI, Vaid RL, Adams AD. Field defectdue to posterior cortical paraxial lens opacity.A Case Report. Trans Ophthalmol Soc UK1978;98:486–9.

6 Lyne AJ, Phillips CI. Visual field defects dueto opacities in the optical media. Br JOphthalmol 1969;53:119–22

7 Hayashi K, Hayashi H, Nakao F, et al.Influence of cataract surgery on automatedperimetry in patients with glaucoma. Am JOphthalmol 2001;132:41–6

8 Lam BL, Alward WLM, Kolder HE. Effect ofcataract on automated perimetry.Ophthalmology 1991;98:1066–70.

Lack of human papillomavirus inpterygium of Chinese patientsfrom TaiwanWe read with interest that Gallagher et al haddemonstrated the association of human pap-illomavirus (HPV) and pterygium by polymer-ase chain reaction (PCR).1 Several hypothesesconcerning the pathogenesis of pterygia havebeen proposed, including exposure to ultra-violet irradiation2 and other environmental

factors, genetic predisposition, and viralinfections.3 The various theories regardingpterygium formation imply that much aboutthe pathogenesis of pterygia remains to beinvestigated.

The involvement of HPV in the genesis ofpterygia is controversial. Some authors havedemonstrated that HPV is present in 24–50%of specimens, whereas others have failed todetect HPV in pterygia.4–7 To help resolve thisdilemma, we evaluated 65 pterygia, 23pinguecula, and 88 normal conjunctiva de-rived from Chinese patients in Taiwan for thepresence of HPV DNA. We used PCR withthree different consensus primer sets—MY09/MY11 (MY), L1C1/L1C2-1 (LC), and GP5/GP(GP).

Material and methodsSamples were obtained from consecutivepatients treated at the ophthalmological clinicof the Taipei Veterans General Hospital. Medi-cal and ophthalmologic histories were re-corded for each patient, a slit lamp microscopeexamination was performed, and pterygiawere photographed before surgery. In eachcase, a specimen of adjacent clinically normalconjunctival tissue (from the 12 o’clockposition of the corneoconjunctival limbus)was obtained. Immediately after surgery,tissue specimens (pterygia, pingueculas, orconjunctival tissues) were stored at −70°C.

DNA preparationThe DNA from specimens was isolated asdescribed previously.8 Briefly, the lysis buffer(10 mM TRIS-HCl; pH 7.5, 1 mM EDTA, pH7.9; 0.5% SDS) and the proteinase K (100µg/ml) were added to the specimens andincubated overnight at 37°C. The standardphenol-chloroform extraction and the etha-nol precipitation were used for DNA purifica-tion and the pelleted DNA was resuspended in50–100 µl of tridistillated sterile water. Todetermine the quality and quantity of the iso-lated DNA, each pelleted DNA sample wasanalysed by electrophoresis on 1% agarosegels stained with ethidium bromide andviewed spectrophotometrically.

PCR analysis for HPVEach amplification reaction was carried out ina total volume of 20 µl overlaid with one dropof mineral oil and contained 10 mM TRIS-HCl(pH 8.3), 50 mM KCl, 0.25 U Taq DNA-polymerase (Perkin-Elmer), and 100–200 ngDNA. The concentration of dNTPs and MgCl2

varied with each set of primers. Each PCRwas carried out in DNA thermal cycler(Perkin-Elmer CETUS DNA Thermal Cycler480) with the first denaturation step at 92°Cfor 4 minutes and the final extension stepat 72°C for 15 minutes. The conditionsand the number of denaturation-annealing-extraction cycles were different with each setof primers.

Table 1 Consensus primer sequences for human papillomavirus DNAdetection

Primer Sequence* (5′-3′)

MY11 GCMCAGGGWCATAAYAATGGMY09 CGTCCMARRGGAWACTGATCL1C1 CGTAAACGTTTTCCCTATTTTTTTL1C2-1 TACCCTAAATACTCTGTATTGGP5 TTTGTTACTGTGGTAGATACGP6 GAAAAATAAACTGTAAATCA

*M=A+C, R=A+G, W=A+T, Y=C+T.

1046 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

To control the quality of the isolated DNAinternally, the 268 bp sequence of β globulingene was amplified using PC04(5’CAACTTCATCCACGTTCACC3’) primersand GH20 (5‘GAAGAGCCAAGGACAGG-TAC3‘) primers9 in the multiplex PCR with theMY, LC, or GP primers. DNA samples extractedfrom cell cultures infected with HPV wereused as a positive control. Each PCR productwas analysed by electrophoresis on 2% agar-ose gels stained with ethidium bromide.

PCR with MY09 and MY11 consensusprimersThe PCR with MY09/MY11 was performed asdescribed previously.10 The PCR methods withthe three different sets of primers weredescribed previously.11 The PCR mixture wascomplemented with 2.5 mM MgCl2, 0.1 mM ofeach dNTP, 0.5 µM MY09 and MY11 primers(Table 1) and 0.3 µM PC04 and GH 20 primers.The DNA amplication was carried out during30 cycles that included denaturation at 92°Cfor 30 seconds, annealing at 53°C for 30seconds, and primer extension at 72°C for 30seconds.

PCR with L1C1, L1C2-1 consensus primersThe PCR with L1C1/L1C2-1 was performed asdescribed previously.10 The PCR mixture wascomplemented with 4 mM MgCl2, 0.2 mM ofeach dNTP, 0.5 µM L1C1, and 0.25 µM L1C1–1primers (Table 1). The DNA amplication wascarried out during 30 cycles that includeddenaturation at 92°C for 30 seconds, anneal-ing at 53°C for 30 seconds, and primer exten-sion at 72°C for 30 seconds.

PCR with GP5, GP6 consensus primersThe PCR with GP5/GP6 was performed asdescribed previously.10 The PCR mixture wascomplemented with 2.5 mM MgCl2, 0.05 mMof each dNTP, 0.5 µM GP5 and GP6 primers(Table 1) and 0.3 µM PC04 and GH 20 primers.The DNA amplication was carried out during40 cycles that included denaturation at 94°Cfor 30 seconds, annealing at 45°C for 30seconds, and primer extension at 72°C for 30seconds.

ResultsThe specimens included 65 conjunctivalpterygia, 23 pingueculas, and 88 normal con-junctivas. Characteristics of patients areshown in Table 2. We were unable to detectany HPV DNA fragments in the 23 specimensof pingueculae, 65 specimens of pterygia, and88 specimens of normal conjunctiva tested.

CommentIt has been proved that HPV possessesoncogenic potential and contributes to thedevelopment of various preneoplastic andneoplastic conditions.12 DNA of many types ofHPV, particularly types 16 and 18, has beendetected in papillomas, dysplasia, and cancersobserved on the eyelids, lacrimal outflowtract, conjunctiva, and cornea.13 14 In thisstudy, three sets of consensus primers, MY, LC,and GP, were used; we were unable to detectHPV in any pterygium, pinguecula, or normalconjunctival specimen from Chinese patientsin Taiwan, where the prevalence of pterygia ishigh.

Three studies have addressed the presenceof HPV DNA in pterygia and all used PCRamplification with a single primer (Table 3).These reports demonstrated big differences infrequencies, from 0% to 100%, and variety ofHPV types (type 6, 11, 16, 18) that could bepossibly explained by the different primersused, the absence of adequate controls, smallsample size (10–50 specimens), and thepossible different frequencies of HPV infec-tion in geographically distinct populations.Confirmatory larger studies in different geo-graphic populations using more efficientprimer(s) are needed to clarify the relationbetween HPV infection and pterygium forma-tion.

The similar controversy occurred in thedetection HPV DNA of malignant epithelialneoplasms of conjunctiva but not squamouscell papilloma of conjunctiva.15 By reviewingthe published data of previous reports, HPVpositive rates in conjunctival papilloma speci-mens were quite consistent, from 44–75% andmost of the HPV types were type 6 and 11 thatwere classified as low risk HPV genotypes.15

But in the case of malignant epithelial

neoplasms of conjunctiva, the frequencies ofHPV detection varies from 0–100% and bothlow risk, HPV-6 and HPV-11, and high risk,HPV-16 and HPV-18, groups were found byvarious molecular techniques.15

Owing to different populations studied andthe absence of a gold standard HPV detectiontechnique and adequate controls in moststudies published to date,1 6 7 16 there aremarked variations in the obtained HPV preva-lence rates in pterygium. Therefore, HPVprobably does not act alone in the develop-ment of pterygium and the exact role of HPVin the pathogenesis of pterygium remainsunclear. The lack of HPV DNA in pterygium inthis study may indicate either the HPV is notassociated with pterygium formation or thatthe technique was not adequate for demon-stration of such an association. Based on ourdata, we suggest that HPV is not a requiredcofactor in the development of pterygia.

Grant support: none.

Commercial relationship: none.

K-H Chen, W-M Hsu, C-C Cheng, Y-S LiDepartment of Ophthalmology, Taipei Veterans

General Hospital, Taipei, Taiwan

K-H Chen, W-M HsuNational Yang-Ming University, Taiwan

K-H Chen, C-C Cheng, Y-S LiDivision of Medical Engineering, National Health

Research Institutes, Taipei, Taiwan

Correspondence to: Dr Wen-Ming Hsu, Departmentof Ophthalmology, Taipei Veterans General

Hospital, #201, Shih-Pai Road, Section II, Taipei,11217 Taiwan; [email protected]


References1 Gallagher MJ, Giannoudis A, Herrington CS,

et al. Human papillomavirus in pterygium. Br JOphthalmol 2001;85:782–4.

2 Cilova-Atanasova B. On the pathogenesis ofpterygium. Folia Med (Plovdiv)1971;13:67–74.

3 Dushku N, Hatcher SL, Albert DM, et al. p53expression and relation to humanpapillomavirus infection in pingueculae,pterygia, and limbal tumors. Arch Ophthalmol1999;117:1593–9.

4 Varinli S, Varinli I, Koksal Erkisi M, et al.Human papillomavirus in pterygium. Cent AfrJ Med 1994;40:24–6.

5 Detorakis ET, Sourvinos G, Spandidos DA.Detection of herpes simplex virus and humanpapilloma virus in ophthalmic pterygium.Cornea 2001;20:164–7.

6 Detorakis ET, Drakonaki EE, Spandidos DA.Molecular genetic alterations and viralpresence in ophthalmic pterygium. Int J MolMed 2000;6:35–41.

7 Dushku N, Hatcher SL, Albert DM, et al. p53expression and relation to humanpapillomavirus infection in pingueculae,pterygia, and limbal tumors. Arch Ophthalmol1999;117:1593–9.

Table 2 Characteristics of patients with pterygia and pinguecula

Pterygium Pinguecula

Patients (M/F) 65 (40/25) 23 (15/8)Age (years, mean (SE)) 63.3 (5.9) (range 55.5–82.3) 58.3 (7.4) (range

44.4–71.2)Medication for conjunctivitis (%) 20 (30.8) 3 (13.0)Duration of lesion (years, mean (SE)) 9.8 (3.7) (range 5.5–21.5) 18.1 (7.9) (range

10.0–28.5)Conjunctivitis history (%) 24 (36.9) 2 (8.7)

Table 3 Literature reports of human papillomavirus detection in pterygia

Authors (year published)No ofspecimens

Pterygium(type) Method/primer Positive rates

HPV types(negative)

Dolmetsch et al (1996)16 16 ? Immunohistochemical stain 100% 16 (6, 11, 18)Dushuku et al (1999)7 13 P + R My09/My11 0 –Detorakis et al (2000)6 50 P + R GP5/GP6 24% 18Gallagher et al (2001)1 10 ? GP5/GP6 50% 6, 11, 16Chen et al (current study) (2002) 65 P MY09/MY11 0 –

L1C1/L1C2-1, GP5/GP6

P = primary; R = recurrent.

PostScript 1047




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

8 Grce M, Husnjak K, Magdic L, et al. Detectionand typing of human papillomaviruses bypolymerase chain reaction in cervical scrapesof Croatian women with abnormal cytology.Eur J Epidemiol 1997;13:645–51.

9 Bell DA, Taylor JA, Paulson DF, et al. Geneticrisk and carcinogen exposure: a commoninherited defect of the carcinogen-metabolismgene glutathione S-transferase M1 (GSTM1)that increases susceptibility to bladder cancer.J Natl Cancer Inst 1993;85:1159–64.

10 Husnjak K, Grce M, Magdic L, et al.Comparison of five different polymerase chainreaction methods for detection of humanpapillomavirus in cervical cell specimens. JVirol Methods 2000;88:125–34.

11 Husnjak K, Grce M, Magdic L, et al.Comparison of five different polymerase chainreaction methods for detection of humanpapillomavirus in cervical cell specimens. JVirol Methods 2000;88:125–34.

12 Zur Hausen H. Papillomavirus infections: amajor cause of human cancers. BiochimBiophys Acta 1996;1288:55–78.

13 McDonnell JM, McDonnell PJ, Stout WC, etal. Human papillomavirus DNA in a recurrentsquamous carcinoma of the eyelid. ArchOphthalmol 1989;107:1631–4.

14 Lauer SA, Malter JS, Meier JR. Humanpapillomavirus type 18 in conjunctivalintraepithelial neoplasia. Am J Ophthalmol1990;110:23–7.

15 Eng HL, Lin TM, Chen SY, et al. Failure todetect human papillomavirus DNA inmalignant epithelial neoplasms of conjunctivaby polymerase chain reaction. Am J ClinPathol 2002;17:429–36.

16 Dolmetsch AM, Alcocer CE, Scull JJ, et al.The presence of human papilloma virus inpterygia (abstract). Invest Ophthalmol Vis Sci1996;37:S43.

Factor V Leiden mutation doesnot correlate with retinalvascular occlusion in whitepatients with Behçet’s diseaseThe factor V Leiden (FV Leiden) mutationcauses resistance to activated protein C bysubstituting the Glu506 residue with arginineat the cleavage site for activated protein C.Heterozygous carriers of the FV Leiden muta-tion have an increased risk of venous throm-bosis between threefold and sevenfold inpopulation based and family studies.1 2

Behçet’s disease is a chronic inflammatorymultisystem disorder that affects youngadults. The principal cause of visual loss in

this disease is recurrent retinal vein occlusionprobably due to a combination of retinal vas-culitis and thrombus formation. Thrombosisin Behçet’s disease carries a poor ocular andsystemic prognosis, so the presence of anidentifiable and significant risk factor couldbe an indicator for anticoagulant treatment.3

Two recent studies have implicated FV Leidenin the pathogenesis of thrombosis in Turkishpatients with Behçet’s disease. In one study,30% of patients with Behçet’s disease compli-cated by thrombosis were heterozygous orhomozygous for factor V Leiden compared to5.9% of factor V Leiden negative patients.4 Inthe second study, factor V Leiden was detectedin 37.5% of patients with Behçet’s disease anda thrombotic history, compared to 9.4% ofnon-thrombotic patients.5 We have previouslyshown in a study of 106 Middle Easternpatients with Behçet’s disease and 120 raciallymatched controls that the prevalence of factorV Leiden was significantly higher amongpatients with ocular inflammation (odds ratio1.67) and was even more prevalent in patientswho had developed retinal vascular occlusivedisease (odds ratio 2.57).6

In this current study we analysed theassociation between factor V Leiden and clini-cal features of Behçet’s disease in whitepatients from the United Kingdom. Theresults show that, unlike Middle EasternBehçet’s disease patients, factor V Leiden wasnot associated with Behçet’s disease in UKpatients.

PatientsDNA samples from 53 white patients withBehçet’s disease were collected from individu-als attending the Behçet’s disease clinic at theMedical Eye Unit, St Thomas’s Hospital, Lon-don. All patients fulfilled the internationalcriteria for Behçet’s disease. Middle Easternand Afro-Caribbean patients were excludedfrom this study. A total of 150 white controlsfrom the London area were provided from ourDNA bank. Patients’ clinical details wererecorded following full systemic and ocularexamination, the diagnosis of retinal veinocclusion being recorded following fluores-cein angiography.

Factor V Leiden analysisHLA typing and detection of the FV Leidenmutation was performed using PCR-SSP as

previously described.6 The results were ana-lysed by generating two by two contingencytables and statistical analysis was performedusing χ2 test.

ResultsFifty three patients (28 males, 25 female)were analysed; 74% (n=39) had oculardisease, 11 had no ocular disease, and forthree patients the ocular disease status wasunknown. Of those patients with oculardisease, 54% (21/39) had retinal vein occlu-sion.

Twenty two of 53 (42%) were HLA-B*51+ ofwhom 3/22 (14%) were B*5108, the remainderbeing B*5101 (Table 1).

Only 2/53 (3.8%) patients in this cohort ofpatients with Behçet’s disease were hetero-zygous for the FV Leiden mutation (Table 1).Both patients were male, and had oculardisease, however only one of these individualshad evidence of retinal occlusion.

CommentThe factor V Leiden mutation has been linkedwith ocular disease in Middle Eastern pa-tients with Behçet’s disease, in particularthose with proved retinal venous thrombosis.The current data on UK patients with Behçet’sdisease do not show a similar association. Theprevalence of FV Leiden in the patient groupwas no different from the control group.Moreover, while both patients positive for FVLeiden had ocular disease this is against abackground of a high level of eye disease inthis group.

There are several possibilities that couldexplain the difference between the twogroups. Firstly, the presence of FV Leiden inthe Middle Eastern population was particu-larly high (17%) and this may have accountedfor the functional role of this molecule in reti-nal venous thrombosis in this ethnic popula-tion. By comparison, the low prevalence of themutation in white people suggests that muchlarger numbers of Behçet’s disease patientswill need to be tested to identify any possibleassociation. This has been supported by stud-ies on other European patients with Behçet’sdisease where FV Leiden was not identified asa risk factor for systemic venous thrombosis.7

Moreover, in our previous study, we identifiedseveral patients who were homozygous for theFV Leiden mutation and were clinically blind.In a population with such a high prevalence ofthe mutation, homozygosity will be morecommon and may have biased the data infavour of an association between FV Leidenand severity of ocular disease in the patientgroup.6

Secondly, recent studies in relatives of indi-viduals with venous thrombosis have shownthat the presence of FV Leiden adds only athreefold risk of thromboembolism. Over halfof these events were linked to other riskfactors such as pregnancy, surgery, or oralcontraceptives.8 This would suggest that in thegeneral white population genetic mutationsaffecting proteins involved in the coagulationcascade might only be associated with throm-bosis in individuals with concurrent risk fac-tors.

Thirdly, population specific phenotypic ef-fects have been described for other gene poly-morphisms. In a worldwide survey of HIV+and HIV individuals, a particular haplotype ofthe RANTES gene was associated with in-creased risk of acquiring HIV-1, and acceler-ated disease progression, in European Ameri-cans, but not African-Americans. A secondRANTES haplotype was associated with de-layed progression of disease in Japanese

Table 1 Clinical and genetic data on 53 white patients with Behçet’sdisease (BD), and 100 healthy white controls

No (%)

BD patients (n=53)Sex Male 28 (53%)

Female 25 (47%)

Ocular disease Yes 39 (74%)No 11 (21%)

HLA-B*51 Positive 22 (42%)Negative 31 (58%)

FV Leiden Positive 2 (3.8%)Negative 51 (96.4%)

Controls (n=100)HLA-B*51 Positive 6 (6%)

Negative 94 (94%)

FV Leiden Positive 5 (5%)Negative 95 (95%)

1048 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

patients, but not in other ethnic groups ofpatients, probably because this haplotype israrely found in non-Far East Asians.9 Thereare several other factor V gene polymorphismsthat may be involved in white patients andthese could be an area for future study.

These results suggest that interindividualand interpopulation specific genotypes areassociated with disease although the pheno-typic outcome remains the same. Thereforegene polymorphisms that associate withdisease in one population cannot be regardedas associating with the disease in differentethnic groups. It may not be possible to iden-tify genes involved in severity of a complexdisease such as Behçet’s disease, which willhold across different patient populations.

Y Chen, M R Stanford, G R WallaceDepartment of Ophthalmology, King’s College,

London SE1 7EH, UK

R W Vaughan, E KondeatisDepartment of Tissue Typing, King’s College,

London SE1 7EH, UK

F FortuneDepartment of Oral Medicine, University of Leeds,

Leeds, UK

Correspondence to: Dr Graham Wallace,Academic Unit of Ophthalmology, University of

Birmingham, City Hospital, Dudley Road,Birmingham B18 7QU, UK;

[email protected]


References1 Svensson PJ, Dahlback B. Resistance to

activated protein C as a basis for venousthrombosis. N Engl J Med 1994;330:517–22.

2 Martinelli I, Mannucci PM, De Stephano V,et al. Different risks of thrombosis in fourcoagulation defects associated with inheritedthrombophilia: a study of 150 families. Blood1998;92:2353–8.

3 Mochizuki M, Akduman L, Nussenblatt RB.Behçet’s disease. In: Pepose JS, Holland GN,Wilhelmus KR, eds. Ocular immunology andinflammation. St Louis: Mosby, 1996:663–75.

4 Oner AF, Gurgey A, Gurler A, et al. Factor VLeiden mutation in patients with Behçet’sdisease. J Rheumatol 1998;25:496–8.

5 Gul A, Ozbek U, Ozturk C, et al.Coagulation factor V gene mutation increasesthe risk of venous thrombosis in Behçet’sdisease. Br J Rheumatol 1996;35:1178–80.

6 Verity DH, Vaughan RW, Madant W. FactorV Leiden mutation is associated with ocularinvolvement in Behçet’s Disease. Am JOphthalmol 1999;128:352–6.

7 Espinosa G, Font J, Tassies D, et al. Vascularinvolvement in Behçet’s disease: relation withthrombophilic factors, coagulation, activationand thrombomodulin. Am J Med2002;112:37–43.

8 Martinelli I, Bucciarelli P, Margaglione M, etal. The risk of venous thromboembolism infamily members with mutations in the genes offactor V or prothrombin or both. Br JHaematol 2000;111:1223.

9 Gonzalez E, Dhada R, Bamshad M, et al.Global survey of genetic variation in CCR5,RANTES, and MIP1α: impact on theepidemiolog yof the HIV pandemic. PNAS2001;98:5199–204.

Localised corneal amyloidosisassociated with herpetic keratitisAmyloid diseases are secondary protein struc-ture diseases in which insoluble protein fibrilsaccumulate extracellularly. Twenty differenttypes of fibrils have been described in humanamyloidosis, each with a different clinical pic-ture. Amyloidosis can be generalised, affectingmultiple organ systems, or localised and can

affect almost any organ of the body. In the eyeamyloid is the material commonly seen in lat-tice corneal dystrophy. Cases of localisedcorneal amyloidosis have been reported in lit-erature but are quite rare.1–4 We report a case oflocalised corneal amyloidosis presenting as alarge raised gelatinous vascularised lesion in apatient with long standing herpetic keratitis.

Case reportA fit and healthy 34 year old woman was a ter-tiary referral to the corneal clinic with a longstanding history of a lesion on her right cornea.The initial presentation as a teenager was of ared sore right eye with a corneal ulcer that wastreated as a bacterial ulcer for a few years andlater on as recurrent herpetic keratouveitis. Shehad had numerous intermittent courses ofcombined topical antivirals and steroids withresolution of symptoms. Over the past 2 yearsshe was noted to develop a raised vascularisedlesion over the right cornea, which gave a con-stant foreign body sensation and occasionalepisodes of pain. It was the appearance of thelesion and the discomfort rather than thereduced visual acuity, which prompted her toseek treatment. On presentation in the clinicshe had a visual acuity of 6/36 right (6/24 withpinhole) and 6/6 left eye. Anterior segmentexamination showed a large, raised, gelatinous,slightly nodular, vascularised lesion on theright cornea (Fig 1). The rest of the anteriorsegment examination was normal. Ocular adn-exae did not show any signs of chronic lid dis-ease. The corneal sensation was intact. Asuperficial keratectomy was performed undergeneral anaesthesia to excise the lesion. His-topathological examination of the specimenrevealed a diagnosis of amyloidosis (Fig 2).

CommentAmyloidosis can be either primary or second-ary, both of which can be further classifiedinto systemic and local disease. Systemic pri-mary amyloidosis can affect various ocular

structures presenting as papules or purpuraon the lids, conjunctival deposits, externalophthalmoplegia, vitreous opacities, and glau-coma. Secondary systemic amyloidosis rarelyaffects the eye, although a case of conjunctivalamyloidosis has been reported in a patientwith rheumatoid arthritis.5

Stafford and Fine, for the first time in 1966,reported a case of corneal amyloidosis in ayoung girl with ocular complications of retin-opathy of prematurity.6 Primary familial amy-loidosis, which presents as nodular whitesubepithelial protuberances in the centralcornea, has been postulated to be autosomalrecessive.1 In secondary localised cornealamyloidosis, the material is deposited as aresult of chronic inflammation and irritationfrom scarred lids from trachoma, trichiasis, orlong standing corneal scars.2–4 7–9 To the best ofour knowledge its association with herpetickeratitis has not been reported.

Macpherson et al retrospectively examined200 specimens of corneas removed for variousreasons specifically for amyloid deposits andfound it present in seven cases (3.5%).7 It hasbeen proposed that the basal cells of thecorneal epithelium are responsible for thesynthesis of amyloid,2 although other sourceshave been also proposed.8

D Tejwani, A Azuara-BlancoDepartment of Ophthalmology Aberdeen Royal

Infirmary, Grampian University Hospitals, UK

J MacKenzieDepartment of Pathology

Correspondence to: Augusto Azuara-Blanco, TheEye Clinic, Aberdeen Royal Infirmary, Aberdeen

AB25 2ZN, UK: [email protected]


References1 Stock EL, Kielar RA. Primary familial

amyloidosis of the cornea. Am J Ophthalmol1976;82:266–71.

2 Kigasawa K, Mashima Y, Yogata T, et al. Ahistopathological study of corneal amyloidosissecondary to trichiasis. Nippon GankaGakkai Zasshi 1996;100:394–400.

3 Watts J, Frank H. Corneal amyloidosis. Br JOphthalmol 1989;73:674–6.

4 Brownstein M, Elliot R. Ophthalmologicaspects of amyloidosis. Am J Ophthalmol1970;69:423–30.

5 Blodi FC Apple DJ. Localised conjunctivalamyloidosis. Am J Ophthalmol1979;88:346–50.

6 Stafford WR, Fine BS. Amyloidosis of thecornea. Report of a case without conjunctivalinvolvement. Arch Ophthalmol1966;75:53–63.

7 McPherson SD Jr, Kiffney GT Jr, Freed CC.Corneal amyloidosis. Am J Ophthalmol1966;62:1025–33.

8 Garner A. Amyloidosis of the cornea. Br JOphthalmol 1969;53:3–81.

9 Ramsey MS, Fine BS, Cohen SW. Localisedcorneal amyloidosis. Case report with electronmicroscopic observations. Am J Ophthalmol1972;73:560–5.

Bilateral macular staphylomas ina patient with cone dystrophyA posterior staphyloma is characterised byscleral ectasia and is pathognomonic forpathological myopia.1 2 Posterior staphylomasare classified in to five types based on theanatomic location.1 Type 1 staphylomas ex-tend from the nasal border of the optic nerveinto the macular region and are the most fre-quent staphyloma seen in myopes.1 Type 2staphylomas are centred on the macula whiletype 3 staphylomas are centred on the optic

Figure 1 Clinical appearance of the lesionon presentation. Note the gelatinousappearance and the marked vascularisation.

Figure 2 Photomicrograph of a Congo redstained section of the lesion showing amyloidin the corneal stroma.

PostScript 1049




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

disc without macular extension.1 Type 4staphylomas are located nasal to the optic discand type 5 staphylomas develop inferior to theoptic disc.1 Staphylomas can be associatedwith multiple complications including retinalpigment epithelial atrophy, lacquer cracks,retinal and subretinal haemorrhages, andchoroidal neovascularisation.1 3

This report describes a patient with undiag-nosed, bilateral type 2 macular staphylomascompounded by cone dystrophy. To the best ofour knowledge, this is a novel association notreported in the literature and with potentialtherapeutic implications.

Case reportA 32 year old white woman presented to theWilmer Ophthalmological Institute, Balti-more, MD, for a second opinion. She reportedhaving progressively worsening vision sincechildhood and was diagnosed with cone-roddystrophy at age 18 by an outside ophthal-mologist. She experienced photophobia bothindoors and outdoors. She denied recentchanges in her vision. Past ocular history was

otherwise significant for a remote history ofcorneal abrasion in the right eye. Past medicalhistory and family history were non-contributory.

On ophthalmological examination, her un-corrected visual acuity was 20/200-2 in theright eye and 20/200-1 in the left eye. Retino-scopic reflex and refraction were variable andsignificant for mild myopia. Refraction did notimprove her vision. There was no relativeafferent pupillary defect and extraocularmovements were normal. There was noevidence of nystagmus. Slit lamp biomicros-copy of the anterior segment was unremark-able. Dilated fundus examination showed atilted optic nerve head in each eye. There werebilateral macular retinal pigment epithelialchanges consisting of a ring of hypopigmen-tation surrounding an area of mildly in-creased pigmentation centrally (Fig 1). Theretinal vessels in each eye appeared to diveinto posterior staphylomas (Fig 1). The

staphylomas were centred around the maculain each eye. The peripheral retina in each eyewas otherwise normal.

Fluorescein angiography demonstratedmottled hyperfluorescence without leakagecorresponding to the retinal pigment epithe-lium (RPE) changes (data not shown). Gold-mann visual fields were remarkable forcentral scotomas in both eyes with peripheralisoptres full to II-4 stimulus in the right eyeand I-4 stimulus in the left eye (Fig 2). AB-scan showed bilateral staphylomas withmacular involvement (Fig 3). On electroretin-ography, photopic responses were markedlyreduced. The dim scotopic responses werenormal. The mixed scotopic responses were90% of normal in the right eye and 97% ofnormal in the left eye. There were markedlyreduced photopic flash and flicker responses,with a questionable response of 10% of thenormal amplitude. Pelli-Robson contrast sen-sitivity testing was depressed at 1.2 log unitsin a dim environment (normal = 1.65). D15colour testing detected four major and threeminor errors in the right eye, and five majorand two minor errors in the left eye. A thera-peutic red tinted contact lens was prescribedto eliminate the photophobia and aversion tolight due to cone dystrophy, and thereby toreduce the level of visual dysfunction.4 5 After1 month of wear, the patient reported being alot more comfortable in bright surroundings.She did not have to squint as much as beforeusing these lenses, was able to sustainprolonged eye contact with other individuals,had improved face recognition and demon-strated improved posture. Visual acuity was20/125 in each eye tested separately and20/80-2 when both eyes were tested together.

CommentIn summary, we have described a patientwhose findings are consistent with a diagno-sis of cone dystrophy compounded by bilateralmacular staphylomas. We believe that thisdoes not represent congenital achromatopsiagiven the absence of nystagmus and thehistory of progressively worsening vision.Although there is a report of familial conedystrophy with bilateral macularcolobomata,6 we are unaware of a case ofbilateral macular staphylomas associated withcone dystrophy. To our knowledge, this caserepresents a previously unreported associ-ation of cone dystrophy with macular staphy-lomas. Awareness of this association willhopefully contribute to proper diagnosis asthis finding had presumably been missed inprevious ophthalmological examinations.

Figure 1 (A) and (B) Bilateral macularstaphylomas in a patient with conedystrophy. There are macular retinal pigmentepithelial changes consistent with conedystrophy. The retinal vessels in both eyesappear to dive posteriorly into staphylomasthat are centred around the macula (type 2staphyloma).

Figure 2 (A) and (B) Goldmann visual fields in both eyes demonstrate central scotomas with peripheral isoptres full to II-4 stimulus in the righteye (bottom right) and I-4 stimulus in the left eye (bottom left).

Figure 3 (A) Horizontal B-scan ultrasoundof the right eye. The depth and width of themacular staphyloma is 1.5 mm and 4.6 mmrespectively. (B) Horizontal B-scan ultrasoundof the left eye. The depth and width of themacular staphyloma is 1.0 mm and 4.2 mmrespectively.

1050 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

Given the significant association of macularstaphylomas with numerous complicationslisted above, especially the risk for choroidalneovascularisation and haemorrhage, suchpatients should receive counselling regardingits symptoms and receive periodic compre-hensive ophthalmological examinations.

Financial interests: None.

Financial support: None.

R S Apte, J S SunnessThe Retinal Vascular Center, The Wilmer

Ophthalmological Institute, The Johns HopkinsUniversity School of Medicine, Baltimore, MD, USA

R S Apte, B G GoldsteinThe Vitreoretinal Division

W L Park, J S SunnessLions Vision Center

R Z Raden, M J ElmanElman Retina Group, PA, Baltimore, MD, USA

Correspondence to: Janet S Sunness, MD, TheWilmer Ophthalmological Institute, The Johns

Hopkins University School of Medicine, 550 NBroadway, 6th Floor, Baltimore, MD 21205, USA;

[email protected]

Accepted for publication 2 January 2003

References1 Quaranta M, Brindeau C, Coscas G, et al.

Multiple choroidal neovascularizations at theborder of a myopic posterior macularstaphyloma. Graefes Arch Clin ExpOphthalmol 2000;238:101–3.

2 Curtin BJ, Carlin DB. Axial lengthmeasurements and fundus changes in themyopic eye. Am J Ophthalmol1971;71:42–53.

3 Steidl SM, Pruett RC. Macular complicationsassociated with posterior staphyloma. Am JOphthalmol 1997;123:181–7.

4 Young RSL, Krefman RA, Fishman GA. Visualimprovements with red-tinted glasses in apatient with cone dystrophy. Arch Ophthalmol1982;100:268–71.

5 Zisman F, Harris MG. Therapeutically tintedcontact lenses. In: Harris MG, London R, eds.Contact lenses: treatment options for oculardisease. St Louis: Mosby, 1996:105–22.

6 Miller SA, Bresnick G. Familial bilateralmacular colobomata. Br J Ophthalmol1978;62:261–4.

Bloody tears, and more! Anunusual case of epistaxisA 56 year old woman presented to theemergency department with a 2 hour historyof bleeding from the right nostril. She hadtried to stop it by pinching her nose but hadthen also experienced bleeding from her righteye and ear (Fig 1).

Bloody tears, or haemolacria, are an occa-sional feature of hereditary haemorrhagictelangiectasia,1 and tumours of the lacrimalapparatus.2 In the emergency department,however, they are more commonly encoun-tered accompanying epistaxis. To date,Medline lists only a single case report ofhaemolacria in this context,3 and the photo-graph presented here may well be the first ofthe phenomenon.

Its anatomical basis lies in the intimateconnection of nose and eye via the lacrimalapparatus. An increase in pressure within thenasal cavity during epistaxis—for example, bypinching or blowing the nose, can causeretrograde flow of blood through the systemand thus lead to bloody tears emerging fromthe ipsilateral eye.

As our patient had longstanding perfora-tion of both tympanic membranes, the bloodin her nose was also able to travel retrogradevia the auditory tube and middle ear into theexternal auditory canal. This led to theadditional bleeding from the right ear.

Bleeding was readily controlled by nasalsponge tamponade. The patient made anuneventful recovery.

M F WieseDepartment of Emergency Medicine, King’s College

Hospital, Denmark Hill, London SE5 9RS, UK;[email protected]


References:1 Soong HK, Pollock DA. Hereditary

hemorrhagic telangiectasia diagnosed by theophthalmologist. Cornea 2000;19:849–50.

2 Levine MR, Dinar Y, Davies R. Malignantmelanoma of the lacrimal sac. OphthalmicSurg Lasers 1996;27:318–20.

3 Banta RG, Seltzer JL. Bloody tears fromepistaxis through the nasolacrimal duct. Am JOphthalmol 1973;75:726–7.

Haemorrhagic toxoplasmicretinochoroiditis: description ofan unusual clinical presentationToxoplasmic retinochoroiditis (TRC) is aninfectious disease caused by the protozoanToxoplasma gondii. This infection affects manyorgans including the eyes. Most of the timeocular involvement occurs after a transplacen-tal transmission, throughout pregnancy, butthe infection can also be acquired. Inimmunocompetent patients, TRC is the mostcommon cause of infection affecting the pos-terior segment. Clinically, the lesion appearsas a white focal necrosis involving the fullthickness of the retina, at the margin of an oldpigmented chorioretinal scar. A vitreous in-flammation is usually present and occasion-ally vasculitis is observed.1

We report the case of a healthy patient whodeveloped a unilateral haemorrhagic retino-choroiditis (RC). The investigations per-formed were positive for a TRC.

Case reportA 45 year old African man was referred with a10 day history of a painless progressive visualloss affecting the left eye. No other ophthal-mological or systemic complaints werepresent. His past medical history was unre-markable.

Ophthalmological examination disclosed avision of 20/20 in the right eye withoutcorrection and in the left eye the best visualacuity was 20/200. Anterior segment exam-ination was normal in the right eye butrevealed a mild inflammation in the left.

Intraocular pressure was within the normallimits in both eyes. Left eye fundus examina-tion showed a vitreous inflammation (cells:++) and a whitish retinochoroidal lesion sur-rounded by a large preretinal haemorrhage.Hard exudates were present in the maculararea as well as a retinal oedema (Fig 1).

Investigations revealed an erythrocyte sedi-mentation rate of 6 mm in the first hour (ref-erence range 1–12), and a normal white bloodcount. Serological testing for toxoplasmosisgave negative results for IgM but IgG titreswere 40 IU/ml (reference range >3). Serologytests for Borrelia burgdorferi, Treponema palli-dum, and HIV were normal. An acute infectionwas suspected and we decided to perform ananterior chamber tap. Polymerase chain reac-tion (PCR) (toxoplasmosis, CMV, HSV, VZV)gave negative results, but the Goldmann-Wittmer coefficient was 13.64 (referencerange <4), revealing a local production ofanti-toxoplasmic immunoglobulins. Tests forsarcoidosis and for connective tissue disorderswere negative. Immunoglobulin electrophore-sis, quantitative immunoglobulin levels, CD4-CD8 lymphocyte count, C3-C4 and CH50examination were within the normal range.PPD skin test was just positive (7 mm). Chestx ray was normal.

Based on these findings, a TRC wasdiagnosed. The patient was treated withsulphadiazine (4 × 1 g/day), pyrimethamine(2 × 25 mg/day) and folinic acid, during 6weeks. Topical steroids and mydriatic dropswere also prescribed. Prednisone (1 mg/kg)was introduced, at tapering doses, during thetreatment.

After 3 months, visual acuity returned to20/20 without a correction in the left eye.Anterior segment examination was normal.Left eye posterior segment examination dis-closed a regression of the haemorrhages and awhite chorioretinal scar with hard exudateslocated around the fovea. Kyrieleis’s plaquewere observed along the inferior papillaryarterial vessel (Fig 2).

The patient was followed during 2 yearsand no reactivation of the RC was observed.Moreover, tests to exclude an immune diseasewere still within the normal limits (HIV,immunoglobulin electrophoresis, quantitativeimmunoglobulin levels, PPD skin test, CD4-CD8 lymphocytes count, C3-C4, and CH50).

CommentThe most classic clinical presentation of anactive toxoplasmic lesion is that of a whitishand oedematous necrotising RC close to anold pigmented scar. A severe vitreous

Figure 1 Right sided haemolacna andbloody otorrhoea in the context of epistaxis.Photograph reproduced with the consent ofthe patient.

Figure 1 Fundus of the left eye. Presence ofan oedematous retinochoroidal lesionsurrounded by a large preretinalhaemorrhage. Hard exudates are observedin the macular area.

PostScript 1051




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

inflammatory reaction is usually associated,appearing as a “headlight in the fog.” Lesionscan occur anywhere in the posterior segmentbut most of the time, they are located in themacular area, affecting one or both eyes.Associated findings include the presence of aninflammatory sheathing of retinal vessels.1

However, a variety of clinical presentationshave been reported in the past; Friedmann etal2 described the presence grey-white finepunctuate lesions affecting the deep retinawith a mild vitreous inflammation. Directoptic nerve involvement by the protozoan wasdescribed by Zimmermann in 1956.3 Morerecently, various clinical aspects of TCR weredescribed in immunocompromised hosts, ap-pearing as diffuse areas of retinal necrosis4 oras a bilateral military retinitis.5

Ocular occlusive vasculitis can be observedin inflammatory diseases including Behçet’ssyndrome, sarcoidosis and systemic lupuserythematosus, in infectious disorders (syphi-lis, acute retinal necrosis), and in TCR. Branchartery obstruction has been reported when avessel passes through an acute TCR,6 6a as wellas retinal vein occlusion.7 8

The case reported here was diagnosticallychallenging as the fundus appearance was notcharacteristic of classic TCR. In fact, haemor-rhages are more frequently found in viralinfections such as cytomegalovirus retinitis.This feature could be explained by damage tothe vascular wall passing through the RC,without signs of a vascular obstruction.

This case demonstrates the importance ofincluding toxoplasmosis in the differentialdiagnosis of unilateral haemorrhagic RC inimmunocompetent patients.

E Baglivo, A B SafranClinique d’Ophtalmologie, Hôpitaux Universitaires

de Genève, Rue Alcide-Jentzer, 22-CH, 05Geneva, Switzerland

Correspondence to: Edoardo Baglivo, MD, Cliniqued’Ophtalmologie, rue Alcide-Jentzer, 22, CH-1205Geneva, Switzerland; [email protected]


References1 Gass DM. Steroscopic atlas of macular

disease: diagnosis and treatment. 3rd ed. Vol2. St Louis: CV Mosby, 1987;464–9.

2 Friedmann CT, Knox DL. Variations inrecurrent active toxoplasmosicretinochoroiditis. Arch Ophthalmol1969;81:481–3.

3 Zimmermann LE. Diseases of the opticnerve: pathology of demyelinating diseases.Trans Am Acad Ophthalmol 1956;60:46–68.

4 Holland GN, Engstrom Re, Glasgow BJ, et al.Ocular toxoplasmosis in patients with theacquired immunodeficiency syndrome. Am JOphthalmol 1988;106:653–67.

5 Berger BB, Egwuagu CE, Freeman W, et al.Miliary toxoplasmosic retinitis in acquiredimmunodeficiency syndrome. ArchOphthalmol 1993;111:373–6.

6 Braunstein RA, Gass JD. Branch arteryobstruction caused by acute toxoplasmosis.Arch Ophthalmol 1980;98:512–13.

6a Fardeau C, Rao NA, Cassoux N, et al.Diagnosis of toxoplasmic retinochoroiditis withatypical clinical features. Am J Ophthalmol2002;134:196–203.

7 Rose GE. Papillitis, retinal neovascularisationand recurrent retinal vein occlusion intoxoplasma retinochoroiditis: a case reportwith uncommon clinical signs. Aust N Z JOphthalmol 1991;19:155–7.

8 Gentile RC, Berinstein DM, Oppenheim R, etal. Retinal vascular occlusions complicatingacute toxoplasmic retinochoroiditis. Can JOphthalmol 1997;32:354–8.

Topical dorzolamide andmetabolic acidosis in a neonateWe describe a neonate with bilateral Peter’sanomaly who became unwell and developed ametabolic acidosis after commencing topicaldorzolamide. He was fully investigated toexclude other causes of acidosis, and subse-quently improved on discontinuation of topi-cal treatment. To the best of our knowledge,there have been no reports of topical carbonicanhydrase inhibitors causing metabolic acido-sis in children or adults.

A 5 day old boy was referred to a tertiarypaediatric ophthalmology unit with bilateralcorneal opacities for consideration of pen-etrating keratoplasty. He had a normal Apgarscore at delivery at 35 weeks’ gestation andweight 2.3 kg. In addition, he had had fullscreening investigations including bloodgases, abdominal ultrasound, and DMSAscans because of a prenatal history of intra-uterine growth retardation with suspicion of asingle kidney.

Ocular examination revealed total leftcorneal opacification and a small opacity ofthe right cornea inferiorly. Intraocular pres-sures measured were normal but since digit-ally the eyes felt firm and applanation tonom-etry is unreliable in thinned corneas, he wascommenced on Trusopt (MSD) eye drops,three times daily to both eyes. He was to bereviewed 2 weeks later.

Seven days following the commencementof Trusopt at his regular paediatric follow upappointment, he was found to be sleepy withpoor feeding and poor capillary refill. Therewas no history of diarrhoea or vomiting.Arterial blood gases revealed a metabolic aci-dosis with a pH 7.08, PCO2 4.2 kp, PO2 11.3 kpand bicarbonate 9.3 mmol/l and base excess of−20.2. There were no markers of infectionwith negative blood, urine, stool, throat, andnasal cultures. Anion gap, serum electrolytes,liver function and urinalysis for pH, specificgravity, and electrolytes were also unremark-able. Renal ultrasound and DMSA scanshowed a normal functioning single rightkidney.

As the cause for the metabolic acidosis atthis stage was unknown he was givenintravenous cefotaxime, flucloxacillin, halfcorrection bicarbonate infusion followed byoral sodium bicarbonate supplements for 3days. He showed some improvement withtreatment; however, he remained significantlyacidotic and unwell. At routine ophthalmicreview 5 days later, while free of all othertreatments, the eye drops were stopped and heshowed spontaneous next-day resolution ofhis acidosis. He symptomatically improvedand gained weight over the subsequent fewdays (Fig 1).

Topical dorzolamide has been shown tocause significant reduction in intraocularpressure (IOP) in children and is welltolerated.1 Secondary glaucoma is well recog-nised in cases of Peter’s anomaly and raisedIOP is well known to cause corneal clouding.2

Congenital corneal opacities necessitate ur-gent treatment in order to reduce amblyopia,and therefore it is essential to excludeglaucoma. Topical Trusopt (MSD) is used rou-tinely at the department of ophthalmology,Great Ormond Street, as it is thought to havelower potential for adverse systemic effectsthan topical β blockers.

Topical dorzolamide is a potent inhibitor ofCA-II1 and this inhibition decreases the rate ofaqueous humour secretion consequently low-ering IOP. In the proximal renal tubule CA-IIis also required to sustain maximal rates ofHCO3 reabsorption. Significant systemic inhi-bition of carbonic anhydrase has not beenobserved and there has been an absence ofdemonstrable metabolic effects in adults.2–6

However, with the oral carbonic anhydraseinhibitor, acetazolamide, the renal carbonicanhydrase involvement and acidosis havebeen shown to be proportionally related to theplasma concentration levels of the drug. Thedose per kg systemic absorption of topically

Figure 2 Fundus of the left eye (3 monthsafter treatment). Regression of thehaemorrhages. Presence of a whitishretinochoroidal scar with hard exudates aroundthe fovea. Kyrieleis’s plaques are observedalong the inferior papillary arterial vessel.

Figure 1 Trend in metabolic acidosis over time.

1052 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

administered dorzolamide would be expectedto be higher in neonates/infants of lower bodyweight compared with adults.

Metabolic acidosis with normal anion gapand serum electrolytes in the absence of diar-rhoea, as in this case, is more likely to be dueto proximal renal tubular bicarbonate loss.Spontaneous improvement of the acidosis ontermination of the topical dorzolamide isstrongly suggestive of the culpability ofdorzolamide. It is unclear as to why this hap-pened, but factors such as prematurity, lowbirth weight, renal tubular immaturity, andone functioning kidney may have led to poorhandling of drug elimination at a higher sys-temic concentration. Although we feel dor-zolamide is a relatively safe topical antihyper-tensive treatment, this case underlines theneed for caution when treating neonates.

S Morris, V Geh, K K NischalDepartment of Ophthalmology, Great Ormond

Street Hospital for Children, London WC1N 3JH,UK

S Morris, V Geh, K K NischalVisual Science Unit, Institute of Child Health,

London, UK

S Sahi, M A S AhmedDepartment of Paediatrics, King George Hospital,

Barley Lane, Ilford, IG3 8YB, UK

Correspondence to: Mr Ken K Nischal, Departmentof Ophthalmology, Great Ormond Street Hospital

for Children, London WC1N 3JH, UK;[email protected]


References1 Portellos M, Buckley EG, Freedman SF.

Topical versus oral carbonic anhydraseinhibitor therapy for pediatric glaucoma.J AAPOS 1998:243–7.

2 DeLuise VP, Anderson DR. Primary infantileglaucoma (congenital glaucoma). Review.Surv Ophthalmol 1983;28:1–19.

3 Sugrue MF. Pharmacological and ocularhypotensive properties of topical carbonicanhydrase inhibitors. Progr Ret Eye Res2000;19:87–112.

4 Strahlman E, Tipping R, Vogel R. TheDorzolamide Dose-response Study Group. Asix-week dose-response study of the ocularhypotensive effect of dorzolamide with aone-year extension. Am J Ophthalmol1996;122:183–94.

5 Biollaz J, Munafo A, Buclin T, et al.Whole-blood pharmacokinetics and metaboliceffects of the topical carbonic anhydraseinhibitor dorzolamide. Eur J Clin Pharmacol1995;47:453–60.

6 Adamsons I, Clineschmidt C, Polis A, et al.The Additivity Study Group. The efficacy andsafety of dorzolamide as adjunctive therapy totimolol maleate gellan solution in patients withelevated intraocular pressure. J Glaucoma1998;7:253–60.

Recurrent infectious crystallinekeratopathy caused by differentorganisms in two successivecorneal grafts in the samepatientInfectious crystalline keratopathy (ICK) is arare complication of penetrating keratoplastycharacterised by an indolent infectious kerati-tis in which needle-like, branching crystallineopacities are seen within the corneal stroma,in the absence of appreciable corneal or ante-rior segment inflammation.1 We report anunusual case of recurrent ICK which occurredin two successive corneal grafts.

Case reportA 63 year old man underwent penetratingkeratoplasty for aphakic bullous keratopathy.The immediate postoperative course wasuneventful. Topical corticosteroid (dexa-methasone 0.1%) was initially given fourtimes daily, and then was tapered to twicedaily. Seven months after transplantation,visual acuity decreased to counting fingerswith no other symptoms. Slit lamp examina-tion showed a focal area of non-suppurativebranching intrastromal white opacities (Fig1). Corneal scrapings for diagnostic smearsand cultures were performed. Microscopicexamination of the smears showed densegroupings of many Gram positive cocci withno inflammatory cells. Cultures grew Strepto-coccus viridans and were negative for fungi. Thepatient was treated hourly with two fortifiedantibiotic eyedrops (amikacin, vancomycin)and topical rifamycin. Topical antibiotictherapy was gradually tapered over 12months. Topical dexamethasone was with-drawn and topical ciclosporin was used tomaintain an immunosuppression. Despiteintensive treatment with appropriate antibi-otics, ICK increased in size and evolved simul-taneously towards abscess and acute rejec-tion. The subsequent corneal condition wassevere residual scarring of the central corneawith diffuse neovascularisation. A secondpenetrating keratoplasty was then performed19 months after the first transplantation.Topical dexamethasone, ciclosporin, and ri-famycin were given four times daily. Threemonths after the second graft, slit lampexamination showed a large central epithelialdefect with multiple diffuse white opacitiesconfined to the anterior stroma. These multi-ple opacities merged into a larger confluentdense opacity near the continuous suture (Fig2). Cytological studies and cultures of the cor-neal scrapings were performed. Light micros-copy disclosed aggregations of many fungiwith no inflammatory cells. Cultures yieldedCandida albicans sensitive to amphotericin Band fluconazole. Bacterial cultures were nega-tive. Topical amphotericin B was started everyhour, along with oral fluconazole. Cortico-steroids were stopped and topical ciclosporinwas maintained. Despite intensive treatment,the infectious keratopathy slowly worsenedover 6 months and the corneal infiltrates werereplaced by scarring and neovascularisation.

CommentTypically, ICK develops in a corneal graft afterlong term use of topical corticosteroids.Chronic topical corticosteroids used to pre-vent graft rejection produces relative immu-nosuppression allowing infection to develop

with little or no inflammation in the cornea.

Gram positive cocci, usually Streptococcus viri-dans, are commonly isolated from ICK lesions,

but other bacteria, fungi, and mixed infec-

tions have been reported.2–5 To the best of our

knowledge, recurrent ICK has never been

reported in two successive corneal grafts and

with two different organisms. Appropriate

laboratory evaluation is therefore necessary to

guide specific antimicrobial therapy. Discon-

tinuation of topical steroids with aggressive

antibiotic therapy may suffice, but continued

infection, vascularisation, or scar formation

may sometimes affect visual acuity and

corneal graft survival. In this case, medical

treatment failed, despite in vitro susceptibility

of micro-organisms to antibiotics and anti-

fungal drugs. Moreover, immunosuppression

(that is, corticosteroids, ciclosporin), neces-

sary to prevent graft rejection, worsened the

infection and did not prevent the acute rejec-

tion process from developing.

In conclusion, this case suggests that local

immunosuppression and factors related to the

patient ocular surface may be predisposing

factors for the development of ICK.

O Touzeau, T Bourcier, V M Borderie,L Laroche

Quinze-Vingts National Center of Ophthalmology,Paris, France

Correspondence to: Tristan Bourcier, MD, PhD,Ophthalmology 5, Quinze-Vingts National Centerof Ophthalmology, 28 rue de Charenton 75012

Paris, France; [email protected]

Accepted for publication 2 February 2003

References1 Sharma N, Vajpayee RB, Pushker N, et al.

Infectious crystalline keratopathy. CLAO J2000;26:40–3.

2 Gorovoy MS, Stern GA, Hood CI, et al.Intrastromal noninflammatory bacterialcolonization of a corneal graft. ArchOphthalmol 1983;101:1749–52.

3 Wilhelmus KR, Robinson NM. Infectiouscrystalline keratopathy caused by Candidaalbicans. Am J Ophthalmol 1991;112:322–5.

4 Rhem MN, Wilhelmus KR, Font RL. Infectiouscrystalline keratopathy caused by Candidaparapsilosis. Cornea 1996;15:543–5.

5 Khater TT, Jones DB, Wilhelmus KR.Infectious crystalline keratopathy caused bygram-negative bacteria. Am J Ophthalmol1997;124:19–23.

Figure 1 Infectious crystalline keratopathydue to Streptococcus viridans in the firstcorneal graft. Except for a small erosion overthe central area of the lesion, the cornealepithelium was intact. There was no corneal,limbal, or anterior chamber inflammation.

Figure 2 Slit lamp photography showing alarge central epithelial defect with multiplediffuse white opacities due to Candidaalbicans in the second corneal graft. Theremaining graft was clear. There were nosigns of ocular inflammation and nosymptoms.

PostScript 1053




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

Rosai Dorfman disease or sinushistiocytosis with massivelymphadenopathy of the orbitSinus histiocytosis with massive lymphaden-opathy (SHML) or Rosai Dorfman syndromeis a rare benign proliferative histiocyticdisease of unknown origin. It predominantlyaffects the lymph nodes. The head and neckregion usually in association with lymph nodeinvolvement, represents one of the most com-mon extranodal areas affected by SHML. Theother common extra nodal site is skin. Rarely,there is widespread dissemination with liver,kidney, respiratory organs, orbit, and eyeballinvolvement.1 The mean age of onset is 20years (birth to 74 years).

Case reportA 57 year old woman with a 6 month historyof double vision was referred to the Royal Vic-toria Eye and Ear Hospital, Dublin. She wasfound to have proptosis, ptosis, diplopia due toinferior rectus dysfunction, and restriction ofelevation of the left eye. Her visual acuity wasnormal. Relevant investigations showed ahigh erythrocyte sedimentation rate (ESR) of44 mm in the first hour, C reactive protein of1.9 (normal less than 1). Her thyroid functiontests, including thyroid microsomal and thy-roglobulin antibodies, were normal. The an-tiacetyl choline receptor antibodies were alsonegative. A computed tomograph (CT) scan ofthe orbit was performed which showed anextraconal soft tissue mass with well definedmargins in the inferomedial part of the leftorbit and no separation from inferior andmedial rectus. There was no bony erosion andthe optic nerve appeared normal. She had anexcision biopsy performed through lateralorbitotomy with Wright’s modification.

The tumour was removed within thecapsule, it was found to be adherent to theinferior and lateral rectus. Histological exam-ination of the tumour revealed an inflamma-tory process composed of aggregates oflymphocytes, with reactive lymphoid follicles,plasma cells, and groups of large histiocyteswith abundant foamy cytoplasm. The inflam-matory process extended around the nerves.There was no vasculitis, areas of coagulativenecrosis, or granuloma formation. The largehistiocytic cells were characterised by round tooval vesicular, hypochromatic nuclei withvesicular chromatin and abundant eosi-nophilic, foamy or clear cytoplasm with poorlydefined cell borders. Emperipolesis waspresent. The phagocytosed cells were mostoften erythrocytes, lymphocytes, and polymor-phonuclear leucocytes (Fig 1). Special stainsfor micro-organisms were negative. Immuno-histochemical stains revealed the presence ofdiffuse S100 positivity within the cells. Thesecells also showed reactivity for the macro-phage marker CD68. The diagnosis of RosaiDorfman disease or SHML was confirmed.

Our patient did not have any lymphaden-opathy or any other extranodal involvement.She did not receive any treatment and after 3years’ follow up there was no sign ofrecurrence. She still had some residual hypo-tropia.

CommentWe report this case to draw attention to thisunusual presentation of SHML confined tothe orbit without any extranodal lesions,which to our knowledge is the only the thirdreported case of this nature. SHML is a rare,benign proliferative histiocytic disease withmassive lymphadenopathy. Table 1 lists thecauses of histiocytic proliferations in the orbit.

In one report of SHML, uveitis withpapilloedema was the only presentationand in another report the only site of thelesion was lacrimal sac with the duct butthese patients later developed cervicallymphadenopathy.2 3 Another case with ocularinvolvement was reported with uveitis andmarginal corneal infiltrates in associationwith cervical lymphadenopathy.4 SHML isusually benign, low grade, and self limitingbut death has been infrequently attributed toit. The condition has also been occasionallyassociated with the development of malignantlymphoma. Hodgkin’s and the follicular typeof non-Hodgkin’s lymphoma and SHML havebeen identified in the same lymph nodebiopsy specimen.5 6 SHML may be associatedwith fever, leucocytosis, elevated erythrocytesedimentation rate, and hypergammaglobuli-naemia. Some studies suggest that humanherpes virus (HHV-6) may play a part inpathogenesis of SHML.7 HHV-6 which infectsmany in childhood and remains latent in hostcells can be reactivated by immunodeficiency.Serological evidence of HHV-6 and Epstein-Barr virus infection have also been docu-mented in patients with SHML with theirpresence in affected tissues as well.7 SHML isusually self limiting but in some cases therewas orbital involvement with compressiveoptic neuropathy, persistent uveitis withmarginal corneal infiltrates, massive lym-phadenopathy impairing cervical perfusion,

and generalised lymphadenopathy with AAamyloidosis. These cases were treated withchemotherapy and oral steroids, the common-est being cyclophosphamide, vincristine, mer-captopurine, and prednisolone. Treatmentcauses regression of the tumour and resolu-tion of cervical lymphadenopathy with mini-mal recurrence.8 9 Our patient did not receiveany treatment and in the 3 year follow upthere was no evidence of recurrence of thedisease in the orbit or any sign of sinus histio-cytosis elsewhere in the body.

R Khan, P MoriartyDepartment of Ophthalmology, Royal Victoria Eye

and Ear Hospital, Dublin, Ireland

S KennedyNational Ophthalmic Pathology Laboratory, Royal

Victoria Eye and Ear Hospital, Dublin, IrelandCorrespondence to: Dr Khan, Department ofOphthalmology, Royal Victoria Eye and Ear

Hospital, Dublin, Ireland

Accepted for publication 9 September 2002

References1 Silvestre JF, Aliga A. Cutaneous sinus

histiocytosis and chronic uveitis. PaediatrDermatol 2000;17:377–80.

2 Dolman PJ, Harris GJ, Weiland LH. Sinushistiocytosis involving the lacrimal sac andduct. Arch Ophthalmol 1992;110:448–9.

3 Pivetti-Pezzi P, Torce C, Collabi-Gisoldi RA.Relapsing bilateral uveitis and papilloedemain sinus histiocytosis with massivelymphadenopathy (Rosai Dorfman syndrome)Eur J Ophthalmol 1995;5:59–62.

4 Rumelt S, Cohen I, Rehany U. Marginalcorneal infiltrates a possible new manifestationof sinus histiocytosis with massivelymphadenopathy. Cornea 2000;19:857–8.

5 Lu D, Estililla OC, Manning JT Jr. Sinushistiocytosis with massive lymphadenopathyand malignant lymphoma involving the samelymphnode. Mod Path 2000;13:414–19.

6 Krezmiecki K, et al. The Rosai Dorfmansyndrome in a 17 year old girl withtransformation to high grade lymphoma. AnnOncol 1996;9:977–9.

7 Levine PH, Jaffe ES. Detection of humanherpes virus 6 it tissues involved by sinushistiocytosis with massive lymphadenopathy. JInfect Dis 1992;166:291–5.

8 Rocken C, Weiker K, Grote HJ, et al. RosaiDorfman syndrome with generalized AAamyloidosis. Hum Pathol 2000;31:621–4.

9 Goldberg S, Mahadevia P, et al. Sinushistiocytosis with massive lymphadenopathyinvolving the orbit: reversal of compressiveoptic neuropathy after chemotherapy. JNeuro-ophthalmol 1998;18:270–5.

CORRECTION

The authors wish to correct an error in thearticle: A Comparison of Perimetric Resultswith Medmont and Humphrey Perimeters.(Br J Ophthalmol 2003:87:690–4). Table 4, row1, column 4 should read 1 not 2, and row 2,column 4 should read 35 not 34. Table 6, row 3,column 1 should read 24 not 27, and row 4,column 2 should read 27 not 24.

NOTICES

Helping the blind and visuallyimpairedThe latest issue of Community Eye Health (No45) discusses help for the blind, with aneditorial by Sir John Wall of the RoyalNational Institute for the Blind on the rightsof blind people. For further information

Figure 1 Sinus histiocytosis. A lymphoidinfiltrate surrounding scattered largehistiocytic cells containing phagocytosedintracytoplasmic lymphocytes (arrows) isseen. Original magnification ×200.

Table 1 Orbital histiocyticproliferations

Secondary to ruptured cyst or traumaSinus histiocytosis with massive lymphadenopathyNecrobiotic xanthogranulomaErdheim-Chester diseaseJuvenile and adult xanthogranulomaLangerhan’s cell histiocytosisFamilial haemophagocytic lymphohistiocytosisTrue histiocytic lymphoma

1054 PostScript




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

please contact: Journal of Community EyeHealth, International Resource Centre, Inter-national Centre for Eye Health, Department ofInfectious and Tropical Diseases, LondonSchool of Hygiene and Tropical Medicine,Keppel Street, London WC1E 7HT, UK (tel:+44 (0)20 7612 7964; email: [email protected]; website: www.jceh.co.uk).Annual subscription (4 issues) UK£28/US$45.Free to developing country applicants.

Second SightSecond Sight, a UK based charity whose aimsare to eliminate the backlog of cataract blindin India by the year 2020 and to establishstrong links between Indian and British oph-thalmologists, is regularly sending volunteersurgeons to India. Details can be found at thecharity’s website (www.secondsight.org.uk)or by contacting Dr Lucy Mathen([email protected]).

SPecific Eye ConditionS (SPECS)SPecific Eye ConditionS (SPECS) is a not forprofit organisation which acts as an umbrellaorganisation for support groups of any condi-tions or syndrome with an integral eye disor-der. SPECS represents over 50 differentorganisations related to eye disorders rangingfrom conditions that are relatively common tovery rare syndromes. The website acts as aportal giving direct access to support groupsown sites. The SPECS web page is a valuableresource for professionals and may also be ofinterest to people with a visual impairment orwho are blind. For further details aboutSPECS contact: Kay Parkinson, SPECS Devel-opment Officer (tel: +44 (0)1803 524238;email: [email protected]; website:www.eyeconditions.org.uk).

The British Retinitis PigmentosaSocietyThe British Retinitis Pigmentosa Society(BRPS) was formed in 1975 to bring togetherpeople with retinitis pigmentosa and theirfamilies. The principle aims of BRPS are toraise funds to support the programme ofmedical research into an eventual cure for thishereditary disease, and through the BRPSwelfare service, help members and their fami-lies cope with the everyday concerns causedby retinitis pigmentosa. Part of the welfareservice is the telephone help line (+44(0)1280 860 363), which is a useful resourcefor any queries or worries relating to theproblems retinitis pigmentosa can bring. Thisservice is especially valuable for those recentlydiagnosed with retinitis pigmentosa, and allcalls are taken in the strictest confidence.Many people with retinitis pigmentosa havefound the Society helpful, providing encour-agement, and support through the Help line,the welfare network and the BRPS branchesthroughout the UK (tel: +44 (0)1280 821 334;email: [email protected]; website:www.brps.demon.co.uk).

Surgical Eye ExpeditionsInternationalVolunteer ophthalmologists in active surgicalpractice are needed to participate in shortterm, sight restoring eye surgery clinics

around the world. Contact: Harry S Brown,Surgical Eye Expeditions International, 27East De La Guerra, C-2, Santa Barbara, CA93101–9858, USA (tel: +805 963 3303; fax:+805 965 3564; email: [email protected] [email protected]; website: www.seeintl.org).

Rise in organ transplantnumbersAccording to UK Transplant, the UK has seenthe highest number of organ transplants insix years. Last year (1 April 2002 to 31 March2003) 2777 patients had their lives saved ordramatically improved through the generosityof 1064 donors. This equated to a 6% increasecompared to the previous 12 months (1 April2001 to 31 March 2002). Furthermore during2002–3, the highest number of people ben-efited from a cornea transplant for five years(1997–98) and 240 more people had theirsight restored than the previous year. For fur-ther information see UK Transplant’s website(www.uktransplant.org.uk).

Elimination of avoidableblindnessThe 56th World Health Assembly (WHA) con-sidered the report on the elimination ofavoidable blindness (doc A56/26) and urgedMember States to: (1) Commit themselves tosupporting the Global Initiative for the Elimi-nation of Avoidable Blindness by setting up anational Vision 2020 plan by 2005; (2) Estab-lish a national coordinating committee forVision 2020, or a national blindness preven-tion committee to help implement the plan;(3) Implement the plan by 2007; (4) Includeeffective monitoring and evaluation of theplan with the aim of showing a reduction inthe magnitude of avoidable blindness by2010; (5) To support the mobilisation ofresources for eliminating avoidable blindness.The WHA also urged the Director-General tomaintain and strengthen WHO’s collabora-tion with Member States and the partners ofthe Global Initiative for the Elimination ofAvoidable Blindness as well as aid in the coor-dination and support of national capability.

MSc course in Community EyeHealthThe International Centre for Eye Health isoffering a full time MSc course in CommunityEye Health from 29 September 2003 to 19September 2004. The course is not clinical andis specifically for eye health professionalswanting to work in the field of community eyehealth. The course is designed in keeping withthe aims, priorities, and strategies of Vision2020—the Right to Sight. The course costs£3939 for home students and £14 110 foroverseas students. Further information: TheRegistry, 50 Bedford Square, LondonWC1B 3DP, UK (tel: +44 (0)20 7927 2239;fax: +44 (0)20 7323 0638; email:[email protected]; website:www.lshtm.ac.uk).

Ophthalmic Anesthesia Society(OAS)—17th Scientific MeetingThe 17th Scientific Meeting of the Ophthal-mic Anesthesia Society (OAS) will be held 3–5

October 2003 at the Westin Michigan AvenueChicago, Chicago, USA. Programme co-chairs:Marc Allen Feldman MD MHS and Steven TCharles MD. The CME joint sponsor is theCleveland Clinic Foundation; CME hours arepending. Fees for OAS members are $300;non-members $475; students $50.Further details: OAS, 793-A Foothill Blvd,PMB 119, San Luis Obispo, CA 93405 USA(tel: +1 805 534 0300; fax: +1 805 534 9030;email: [email protected]; website:www.eyeanesthesia.org).

Glaucoma Society 24th AnnualMeeting and DinnerThe Glaucoma Society 24th Annual Meetingand Dinner will take place on 20 November2003, from 8:30 am to 5:00 pm at The RoyalCollege of Physicians, London, UK. Furtherdetails: Ms Janet Flowers (email:[email protected]).

Detachment Course withinternational faculty on: Retinaland Vitreous Surgery with CasePresentations preceding theAnnual Meeting of IranianSociety of OphthalmologyThe detachment course with internationalfaculty on: Retinal and Vitreous Surgery withCase Presentations preceding Annual Meetingof Iranian Society of Ophthalmology will beheld on 29–30 November 2003 and 1–4December 2003 respectively, at the RaziConference Center, Hemmat Hyw, Tehran,Iran. Further details: Scientific programme:Prof Ingrid Kreissig, University of Tuebingen,Schleichstr. 12, Breuningerbau, 72076 Tuebin-gen, Germany (tel: +49 7071 295209; email:[email protected]).Local organisation: Dr Arman Masheyekhi,Dr Siamak Moradian, Dept of Ophthalmology,Labbanfinejad Medical Center, Pasdaran Ave,Boostan 9, Tehran, 16666, Iran (fax: +98 21254 9039; email: [email protected]).

5th International Symposium onOcular Pharmacology andTherapeutics (ISOPT)The 5th International Symposium on OcularPharmacology and Therapeutics (ISOPT) willtake place 11–14 March 2004, in Monte Carlo,Monaco. Please visit our website for details ofthe scientific programme, registration, andaccommodation. To receive a copy of the Callfor Abstracts and registration brochure pleasesubmit your full mailing details to http://www.kenes.com/isopt/interest.htm.Further details: ISOPT Secretariat (website:www.kenes.com/isopt).

XVth Meeting of theInternationalNeuro-Ophthalmology SocietyThe XVth Meeting of the International Neuro-Ophthalmology Society will take place 18–22July 2004, in Geneva, Switzerland.Further details: Prof. A Safran, UniversityHospital Geneva, c/o SYMPORG SA, Geneva(fax: +4122 839 8484; email: [email protected]; website: www.symporg.ch).

PostScript 1055




j.com/

Br J O


ownloaded from

http://bjo.bmj.com/

PostScript - British Journal of Ophthalmologytive best corrected visual acuity (BCVA) was 3/200...

Documents

Transcript of PostScript - British Journal of Ophthalmologytive best corrected visual acuity (BCVA) was 3/200...