PostScript - British Journal of Ophthalmology · adnexa. Int Ophthalmol Clin 1997;37:39–59. 3 Knowles DM, Jakobiek FA, McNally L, et al. Lymphoid lyperplasia and malignant lymphoma

Subconjunctival corticosteroidsfor benign lymphoid hyperplasiaAppropriate treatment of subconjunctivalbenign lymphoid hyperplasia (BLH) has beenunclear. Most have noted poor response tooral or topical corticosteroids. Many recom-mend observation. Radiotherapy has beenused, but there are risks of vision loss.In this case, we found dramatic response to

local subconjunctival injection of long actingcorticosteroids, which may represent a ther-apeutic option for subconjunctival benignlymphoid hyperplasia.

Case reportA 72 year old woman noticed an enlargingmass on the nasal conjunctiva over the

previous year. She denied any discomfort,diplopia, or visual changes. She was pseudo-phakic in both eyes, and did not have anypast ocular trauma, infection, or eye disease.She had a history of hypothyroidism, multi-ple myeloma, tuberculosis, and pulmonarysarcoidosis. Two months before presentationshe was found to have colon carcinomatreated with colectomy and chemotherapy.She was reportedly free of any tumourmetastasis.Her visual acuity was 20/30 in the right eye

and 20/40 in the left, and extraocular motilitywas full in both eyes. Examination revealed asalmon coloured, raised, and moderately firmpatch on the nasal conjunctiva of the righteye without significant neovascularisation(fig 1A). Dilated fundus examination wasunremarkable. A head MRI scan showedmoderate enlargement of the medial rectusmuscle without involvement of the muscletendon. A simple biopsy (2 mm62 mm61 mm) of the lesion was performed.Pathological examination revealed benign

lymphohistiocytic infiltrates (fig 2). Thelymphoid reaction showed a predominanceof T cells (CD3+), numerous histiocytes, anda smaller number of B cells (CD20+). Therewas no evidence of neoplastic plasma cells,metastatic carcinoma, or well defined sarcoidgranulomas.The patient elected to have surgical treat-

ment over other options including observa-tion. The patient received a 20 mg/0.5 mlsubconjunctival injection of triamcinolone(in 0.5 ml in the nasal bulbar conjunctiva)just superior to the mass. On follow upexamination 2 months later, the patientcontinued to deny any discomfort or visualchanges and was very pleased about herresponse to the treatment. Remarkably, thelesion completely resolved being no longervisible or palpable (fig 1B). This patientunfortunately died 9 months later fromcomplications secondary to a fall, but duringthis time there was no recurrence of theconjunctival lesion.

CommentA patient presenting with a slow growingsalmon coloured subconjunctival massshould always raise suspicion of neoplastic

causes. Patients with ocular BLH and lym-phoma will often have the similar presentingsymptoms and demographic profiles. Inaddition they appear very similar radiologi-cally,1 and thus definitive diagnosis requirestissue biopsy. A pathological diagnosis of BLHtraditionally requires reactive follicles, poly-clonality, and the absence of cytologicalatypia.2 Lymphoproliferative lesions canoccur throughout the ocular adnexa, andsome studies suggest a more benign coursefor conjunctival BLH compared to those inthe orbit.3 Coupland et al found that of 112cases, 32 (29%) were in the conjunctiva, 52(46%) in the orbit and the remainder in theeyelid, lacrimal gland, and caruncle.4 Theoptimal treatment for BLH is uncertain.Many recommend frequent observation.Others have tried focal radiotherapy withsome success,5 but there is a significant riskof vision loss.6 In a recent review of 117 casesof conjunctival lymphoproliferative lesions,17% were BLH, 22% were atypical lymphoidhyperplasia, and 56% were lymphoma.7 Inthese cases 9% were observed, 42% hadcomplete excisional biopsy, 4% had biopsyand cryotherapy, 44% had biopsy and exter-nal beam irradiation, and 6% had biopsy andchemotherapy.7

In this case report, we found a dramaticresponse to local subconjunctival injection ofa long acting corticosteroid. The corticoster-oid near the reactive follicle must have beensufficient to suppress lymphocyte prolifera-tion. This response may represent a thera-peutic option for BLH.

D G Telander, T Z Lee, S E Pambuccian,A J W Huang

Department of Ophthalmology, University ofMinnesota, Minneapolis, MN, USA

S E PambuccianDepartment of Pathology, University of Minnesota,

Minneapolis, MN, USA

D G TelanderUCLA/Jules Stein Eye Institute, Los Angeles, CA, USA

T Z LeeApple Hill Eye Center, 25 Monument Road, Suite 297,

York, PA 17403, USA

Correspondence to: Andrew J W Huang, MD,Department of Ophthalmology, University of

Minnesota, 516 Delaware Street SE, 9th Floor,Minneapolis, MN 55455, USA; [email protected]

doi: 10.1136/bjo.2004.051342

References

1 Knowles DM, Jakobeic FA. Malignant lymphomaand lymphoid hyperplasia occurring in the ocularadnexa. In: Knowles DA, ed. Neoplastichematology. Baltimore: Williams & Wilkins,1992:1009–46.

2 Cockerham GC, Jakobiek FA.Lymphoproliferative disorders of the ocularadnexa. Int Ophthalmol Clin 1997;37:39–59.

3 Knowles DM, Jakobiek FA, McNally L, et al.Lymphoid lyperplasia and malignant lymphomaoccurring in the ocular adnexa (orbit, conjunctiva,and eyelids): a prospective multiparametricanalysis of 108 cases during 1977 to 1987. HumPathol 1990;21:959–73.

Figure 1 (A) Slit lamp photograph of thepatient’s right eye on initial presentation. Notethe size, salmon colour and raised appearanceof the lesion. (B) Slit lamp photograph of thesame lesion 2 months later followingsubconjunctival injection of triamcinolone. Thesubconjunctival lesion appeared to havecompletely resolved with no obvious remnantsseen on the sclera or conjunctiva. (Reproducedwith permission.)

Figure 2 Haematoxylin and eosin staining,106magnification of the lesion biopsied infigure 1A. Note the abundance of lymphocytesseen more clearly in the magnified section in thelower right part of the figure. Also note thepredominance of T cells (CD 3+) with a lessnumerous population of B cells (CD20+) typicalof benign lymphoid hyperplasia.

LETTERS

If you have a burning desire to respondto a paper published in BJO, why not makeuse of our ‘‘rapid response’’ option?

Log onto our website (www.bjophthalmol.com), find the paper that interests you, andsend your response via email by clicking onthe ‘‘eLetters’’ option in the box at the topright hand corner.

Providing it isn’t libellous or obscene, itwill be posted within seven days. You canretrieve it by clicking on ‘‘read eLetters’’ onour homepage.

The editors will decide as before whetherto also publish it in a futher paper issue.

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

Accepted for publication 1 October 2004

770 Br J Ophthalmol 2005;89:770–787

www.bjophthalmol.com

copyright. on N

ovember 10, 2020 by guest. P

rotected byhttp://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.2005.068544/10.1136/bjo.2005.069757 on 27 M

ay 2005. Dow

nloaded from

http://bjo.bmj.com/

4 Coupland SE, Krause L, Delecluse H-J, et al.Lymphoproliferative lesions of the ocularadenexa. Analysis of 112 cases. Ophthalmology1998;105:1430–41.

5 Keleti D, Flickinger JC, Hobson SR, et al.Radiotherapy of lymphoproliferative diseases ofthe orbit. Surveillance of 65 cases. Am J ClinOncol 1992;15:422–7.

6 Bessel EM, Henk JM, Whitelocke RAF, et al.Ocular morbidity after radiotherapy of orbitaland conjunctival lymphoma. Eye 1987;1:90–96.

7 Shields CL, Shields JA, Carvalho C, et al.Conjunctival lymphoid tumors. clinical analysis of117 cases and relationship to systemiclymphoma. Ophthalmology 2001;108:979–84.

Congenital upper eyelid eversioncomplicated by cornealperforationCongenital upper eyelid eversion is a rarecondition more frequently seen in blackinfants and in Down’s syndrome. If recog-nised early, the condition can be managedconservatively without recourse to surgery.We highlight a case that presented late withsevere sight threatening complications.Surgical intervention was consequently theonly appropriate way to manage the patient.

Case reportA 7 month old black female was referred tothe emergency room of the King Khaled EyeSpecialist Hospital in Riyadh, Saudi Arabia.She had a history of a white spot in the rightcornea, and a right upper lid that had flippedup since birth.She was born by breech delivery and was

subsequently diagnosed as having Down’ssyndrome. On examination she had completeeversion of the right upper lid with mildconjunctival chemosis. The lid could berepositioned with ease but when the childcried it reverted. There was a corneal opacitywith a central descematocele and iris adhe-sion to the endothelium. The child wasadmitted and started on intensive topicalantibiotics. In spite of manual inversion ofthe upper lid as well as a moist chamber, thelid continued to re-evert. A lateral tarsor-rhaphy and penetrating keratoplasty wereperformed under general anesthesia but,postoperatively, the eyelid continued to evert(fig 1). Because of the risk of exposure to thetransplanted cornea, a full thickness skingraft to the right upper lid was performed(fig 2). Subsequently, the lid retained itsnormal position and the child could fix andfollow adequately with the eye. The trans-planted cornea was clear at the 10 weekpostoperative visit after which the patientwas lost to follow up.

CommentCongenital eversion of the upper eyelid wasfirst described in 1896 by Adams1 who calledthe condition ‘‘double congenital ectropion.’’The exact incidence of this condition is notknown. Sellar2 reviewed the literature in 1992and found 51 reported cases. Since then onlytwo more case reports could be found in theliterature.3 4 The eversion is usually present atbirth, but late onset of total eversion of theupper eyelids has been described in infancy,5

and as late as 11 years of age.2

The condition typically is bilateral andasymmetrical but unilateral cases have beendescribed. The underlying pathophysiology isobscure and several possible mechanismshave been proposed and associations recog-nised. The incidence appears to be higher inblack infants,6 infants with trisomy 21,2 andin infants born with collodion skin disease.7

Abnormalities such as orbicularis hypotonia,8

birth trauma, vertical shortening of theanterior lamellar or vertical elongation ofthe posterior lamellar of the eyelid and failureof the orbital septum to fuse with the levatoraponeurosis (with adipose tissue interposi-tion),9 absence of effective lateral canthalligament, and lateral elongation of the eyelidhave all been implicated as possible patho-physiological factors. Once everted, orbicu-laris spasm10 may act as a sphincter that leadsto a vicious cycle of conjunctival strangula-tion and oedema secondary to venous stasis.This chemotic conjunctiva usually protectsthe cornea from exposure and, hence, cornealcomplications have thus far not beenreported in infants. To our knowledge thisis the first reported case of congenital eyelideversion complicated by corneal exposure,descematocele, and perforation.Congenital eyelid eversion either resolves

spontaneously with conservative treatment orwith surgical intervention such as a subcon-junctival injection of hyaluronic acid, tarsor-rhaphy with excision of redundantconjunctiva, fornix sutures, and full thick-ness skingraft to the upperlid. In this patientbecause of failure of all conservative mea-sures, the chronicity (7 months) and severity(corneal ulcer/perforation), we elected to doanterior lamellar lengthening by means of afull thickness upper lid skin graft. Cornealperforation necessitating penetrating kerato-plasty is a rare complication of congenitaleyelid eversion, but in this case it wasessential to maintain the integrity of theglobe and to, hopefully, retain vision andprevent deep amblyopia.The condition in this reported case of

congenital eyelid eversion complicated by acorneal ulcer, large descematocele and per-foration, is rare. Congenital upper lid eversion

may present ‘‘once in a lifetime’’ to theophthalmologist or newborn nursery paedia-trician. It is nevertheless important to createawareness among healthcare professionals inobstetric and neonatal care of the existence ofthis potentially sight threatening congenitalanomaly, as the condition is very amenable toearly treatment.

H Al-Hussain, A A Al-Rajhi, S Al-Qahtani,D Meyer

University of Stellenbosch/King Khaled Eye SpecialistHospital, Faculty of Health Sciences, PO Box 19059,

Tygerberg, South Africa

Correspondence to: Professor David Meyer, Universityof Stellenbosch, Faculty of Health Sciences, PO Box19059, Tygerberg, South Africa; [email protected]

doi: 10.1136/bjo.2004.053348

References

1 Adams AL. A case of double congenitalectropion. Med Fortnightly 1896;9:137–8.

2 Sellar PW, Bryars JH, Archer DB. Latepresentation of congenital ectropion of the eyelidsin a child with Down’s syndrome: a case reportand review of the literature. J Pediatr OphthalmolStrabismus 1992;29:64–7.

3 Watts MT, Dapling RB. Congenital eversion of theupper eyelid: a case report. Ophthal PlastReconstr Surg 1995;11:293–5.

4 Dawodu OA. Total eversion of the upper eyelidsin a newborn. Niger Postgrad Med J2001;8:145–7.

5 Siverstone B, Hirsch I, Sternberg MD, et al. Lateonset of total eversion of upper eyelids. AnnOphthalmol 1982;14:477–8.

6 Lu LW, Bansal RK, Katzman B. Primary congenitaleversion of the upper lids. J Pediatr OphthalmolStrabismus 1979;16:149–51.

7 Shapiro RD, Soentgen ML. Collodion skin diseaseand everted eyelids. Postgrad Med1969;45:216–19.

8 Loeffler M, Hornblass A. Surgical management ofcongenital upper-eyelid eversion. OphthalmicSurg 1990;21:736.

9 Blechman B, Isenberg S. An anatomical etiologyof congenital eyelid eversion. Ophthalmic Surg1984;15:111–13.

10 Raab EL, Saphir RL. Congenital eyelid eversionwith orbicularis spasm. J Pediatr OphthalStrabismus 1985;22:125–8.

Homozygous mutation (L527R) ofTGFBI in an individual with latticecorneal dystrophyLattice corneal dystrophy (LCD), an inheritedform of amyloidosis, is characterised by thedevelopment of lattice lines and opacity inthe cornea. LCD is classified clinically intofour subtypes: I, II, III, and IIIA. Severaldistinct mutations of TGFBI have beenassociated with LCDIIIA: P501T,1 L527R,2

N544S,3 A546T,4 N622K (T1913G andT1913A), and V627S.5 All cases of LCDcharacterised at the molecular genetic levelto date have been attributed to heterozygouspoint mutations of TGFBI. We now presentthe first example of a homozygous pointmutation of TGFBI in an individual with LCD,a diagnosis supported by clinical, histological,and molecular genetic findings.

Case reportA 52 year old Japanese man visited ourcorneal clinic in July 1997 with a maincomplaint of gradual impairment of vision.His parents, who were related, were no longeralive and he had no children. He had twobrothers and four sisters. His reporting

Figure 2 Full thickness skin graft to the upperlid successfully expanded the anterior lamella topermanently solve the problem.

Figure 1 After the tectonic penetratingkeratoplasty was performed, the lateraltemporal tarsorrhaphy failed to retain the lid inan inverted position.


PostScript 771


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

suggested that his father had had LCD andthat his younger sister had also developedthis condition.Slit lamp examination revealed a large

region of opacity and thick lattice lines inthe middle to deep portion of the cornealstroma in both eyes (fig 1A and B). Giventhat his visual acuity in both eyes decreasedto 20/40, we performed penetrating kerato-plasty on his left eye in December 2000 andon his right eye in May 2002. After thesurgeries, his visual acuity improved to 20/25in each eye and no recurrence has beenobserved to date.Histological analysis revealed that the

amorphous component of the middle to deepregion of the stroma of both corneas stainedwith eosin (fig 1C) and with Congo red(fig 1D). The detection of apple greendichroism by polarised light microscopy wasalso consistent with amyloid deposition inthe middle to deep region of the corneal

stroma (fig 1E). Although most of theepithelial cell layer, basement membrane,and hemidesmosomes of the right corneaappeared normal by transmission electronmicroscopy (TEM) (fig 1F), a region wasdetected that seemed to be devoid of basalcells, basement membrane, and hemidesmo-somes. TEM also revealed amyloid deposits inthe middle to deep corneal stroma (fig 1G),although Descemet’s membrane andendothelial cells appeared normal.After obtaining informed consent, we

purified DNA from the white blood cellsisolated from 10 ml of the patient’s blood.With appropriate primers,6 we amplifiedexons 4 and 12 of TGFBI by the polymerasechain reaction (PCR) and directly sequencedthe products. We detected a homozygouspoint mutation, CTGRCGG (L527R), incodon 527 of TGFBI in the proband (fig 2).Point mutations were not detected in codons124, 518, 544, 546, or 555, mutations in

which have been associated with cornealdystrophies.

CommentFujiki et al2 reported that LCDIIIA in L527Rheterozygotes is characterised by a late onsetand mild clinical findings. Hirano et al7

reported that the condition caused by L527Rheterozygosity was associated with amyloiddeposition in the deep corneal stroma but notwith corneal erosion. In contrast, LCD in ourpatient with a homozygous L527R mutationwas characterised by onset in middle age,recurrent corneal erosion, and amyloiddeposition in the middle to deep region ofthe stroma. The deposits in the probandshowed right eye-left eye asymmetry in sizeand shape, as previously described for someL527R heterozygotes.2 7

In general, corneal dystrophies caused byhomozygous point mutations in TGFBI arecharacterised by an earlier onset, more severesymptoms, and a higher frequency of recur-rence after keratoplasty compared with thoseattributable to the corresponding heterozy-gous mutations.8–12 Our study and the twoprevious studies2 7 of this mutation suggestthat this is also the case for L527R. However,the difference in the findings of slit lampexamination between L527R heterozygotesand homozygotes with LCD appears to be lessmarked than that observed between R124Hheterozygotes and homozygotes withAvellino corneal dystrophy. The reason forthis discrepancy remains unclear at present.

AcknowledgementsThis research was supported by grants from theJapan Society for the Promotion of Science (JSPS)and the Ministry of Education, Culture, Sports,Science, and Technology of Japan (to MI and TN).

N Yamada, T-i Chikama, N Morishige,R Yanai, T Nishida

Department of Biomolecular Recognition andOphthalmology, Yamaguchi University School of

Medicine, Ube City, Yamaguchi 755-8505, Japan

N Yamada, R Yanai, M InuiDepartment of Pharmacology, Yamaguchi UniversitySchool of Medicine, Ube City, Yamaguchi 755-8505,

Japan

K SekiDepartment of Ocular Pathophysiology, YamaguchiUniversity School of Medicine, Ube City, Yamaguchi

755-8505, Japan

Figure 1 (A and B) Slit lamp photographs of the right and left eyes, respectively, of the patient.Light microscopic (C–E) and transmission electron microscopy (TEM) (F and G) analysis of thesurgically removed corneal specimens of the patient. (C) Haematoxylin and eosin staining of the leftcornea. (D) Congo red staining of the left cornea. (E) Polarised light microscopy of the left cornea.(F) TEM of the epithelial cell layer of the right cornea. (G) TEM of the stromal and endothelial celllayers of the right cornea. Scale bars: 250 mm (C–E) or 10 mm (F and G).

Figure 2 Molecular genetic analysis of TGFBIof the patient. Direct sequencing of PCRproducts corresponding to exon 12 of TGFBI. Ahomozygous TRG mutation (arrow) wasdetected at codon 527.

772 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

Correspondence to: Naoyuki Yamada, MD, PhD,Department of Biomolecular Recognition and

Ophthalmology, Yamaguchi University School ofMedicine, 1-1-1 Minami Kogushi, Ube City,

Yamaguchi 755-8505, Japan;[email protected]

doi: 10.1136/bjo.2004.056168

References

1 Yamamoto S, Okada M, Tsujikawa M, et al. Akerato-epithelin (betaig-h3) mutation in latticecorneal dystrophy type IIIA. Am J Hum Genet1998;62:719–22.

2 Fujiki K, Hotta Y, Nakayasu K, et al. A new L527Rmutation of the betaIGH3 gene in patients withlattice corneal dystrophy with deep stromalopacities. Hum Genet 1998;103:286–9.

3 Mashima Y, Yamamoto S, Inoue Y, et al.Association of autosomal dominantly inheritedcorneal dystrophies with BIGH3 gene mutations inJapan. Am J Ophthalmol 2000;130:516–17.

4 Dighiero P, Drunat S, Ellies P, et al. A newmutation (A546T) of the betaig-h3 generesponsible for a French lattice corneal dystrophytype IIIA. Am J Ophthalmol 2000;129:248–51.

5 Munier FL, Frueh BE, Othenin-Girard P, et al.BIGH3 mutation spectrum in corneal dystrophies.Invest Ophthalmol Vis Sci 2002;43:949–54.

6 Munier FL, Korvatska E, Djemai A, et al. Kerato-epithelin mutations in four 5q31-linked cornealdystrophies. Nat Genet 1997;15:247–51.

7 Hirano K, Hotta Y, Nakamura M, et al. Late-onsetform of lattice corneal dystrophy caused byLeu527Arg mutation of the TGFBI gene. Cornea2001;20:525–9.

8 Mashima Y, Konishi M, Nakamura Y, et al.Severe form of juvenile corneal stromal dystrophywith homozygous R124H mutation in thekeratoepithelin gene in five Japanese patients.Br J Ophthalmol 1998;82:1280–4.

9 Okada M, Yamamoto S, Watanabe H, et al.Granular corneal dystrophy with homozygousmutations in the kerato-epithelin gene.Am J Ophthalmol 1998;126:169–76.

10 Okada M, Yamamoto S, Inoue Y, et al. Severecorneal dystrophy phenotype caused byhomozygous R124H keratoepithelin mutations.Invest Ophthalmol Vis Sci 1998;39:1947–53.

11 Fujiki K, Hotta Y, Nakayasu K, et al. Homozygoticpatient with betaig-h3 gene mutation in granulardystrophy. Cornea 1998;17:288–92.

12 Inoue T, Watanabe H, Yamamoto S, et al.Different recurrence patterns afterphototherapeutic keratectomy in the cornealdystrophy resulting from homozygous andheterozygous R124H BIG-H3 mutation.Am J Ophthalmol 2001;132:255–7.

Vitreous amyloidosis in alanine71 transthyretin mutationFamilial amyloid polyneuropathy (FAP) asso-ciated with mutations in the transthyretin(TTR) gene is the commonest form ofhereditary amyloidosis. The incidence ofvitreous opacities in FAP varies from 5.4%to 35%,1 2 but vitreous opacities as part ofsystemic amyloidosis are virtually pathogno-monic of FAP. Hereditary non-neuropathicsystemic amyloidosis is associated withmutations in the genes for lysosyme, apoli-poprotein A-I, or fibrinogen A a-chain. Thereare some 80 known mutations in TTR gene ofwhich the methionine 30 variant is the mostcommon.3 The rare alanine 71 (Ala 71)variant with vitreous opacities has beendescribed in one family from France andanother from Spain.4 5 We report a case ofFAP Ala 71 without a family history of thedisease who presented with a monocularinferior visual field defect and a correspond-ing vitreous opacity. Amyloid deposition was

subsequently diagnosed on vitreous and suralnerve biopsy.

Case reportA 46 year old woman presented with aninferior visual defect in her left eye. Ocularand systemic evaluation was normal includ-ing brain computed tomography scan.Eighteen months later, the patient developedbilateral floaters and visual loss in the left eyereducing her vision to 20/15 right and 20/30left. Deposits of white ‘‘fluffy’’ material werenoted on the posterior capsule of the left eye(fig 1A) as well as bilateral ‘‘branching’’vitreous opacities, peripheral retinal haemor-rhages and perivascular sheathing (fig 1B).The left eye had a partial posterior vitreousdetachment (PVD), a large vitreous floaterand old inferior vitreous haemorrhage (VH).Vitreous opacification progressed and

visual acuity was reduced to hand movement1 year later. A full blood count, coagulationscreen, and biochemical profile were normal.Creatinine clearance demonstrated a mildreduction in renal function and plasma celldyscrasia was ruled out. Pars plana vitrect-omy successfully cleared the vitreous debris,restoring vision to 20/20 in the left eye.Eighteen months later the patient developeda right foot drop, progressive lower limbnumbness, and numbness in both hands. Asural nerve biopsy established the diagnosisof amyloidosis and immunohistochemistryconfirmed that TTR was the major proteinconstituent of the deposits. Cardiac involve-ment was demonstrated on echocardiographyand renal involvement was confirmed byserum amyloid P (SAP) scintigraphy.Sequencing of her TTR gene confirmed thatshe was heterozygous for the amyloidogenicAla 71 variant.

The patient’s right visual acuity deterio-rated to 20/120 because of increasing whitevitreous opacities and nodular opacities onthe anterior vitreous face. Right vitrectomyresulted in a return of visual acuity to 20/20.An undiluted right vitreous biopsy confirmedlarge amounts of amyloid of the TTR type(fig 2). The patient currently awaits orthopticliver transplantation (OLT).

CommentVitreous opacification was initially attributedto old VH secondary to idiopathic retinalvasculitis. Vitreous biopsy subsequently con-firmed amyloid in the fellow eye. VH mayoccur secondary to vascular adventitial amy-loid deposition or vitreous separation leadingto a retinal tear, although vitreous opacitiesmay be misinterpreted as Kantarjian and deJong first reported vitreous amyloid in FAP.6

Amyloid of the vitreous body has beendescribed as ‘‘glass wool, sheet-like veils orstring of pearls white opacities,’’ whichdiffers from localised ocular amyloid in theorbit, lacrimal gland, conjunctiva, eyelids,sclera, and more specific forms in the cornea.Previous studies have reported abnormalconjunctival vessels, pupillary abnormalities,keratoconjunctivitis sicca, glaucoma, andvitreous opacities.Almost all of the circulating TTR is

produced in the liver and OLT can halt theprogression of this disease and lead to clinicalimprovement.7 SAP scintigraphy is a methodfor identifying and quantitatively monitoringamyloid deposits in vivo,7 but this techniqueis not sensitive enough to monitor vitreousamyloid. Surprisingly, progressive vitreousamyloid deposition has been reported follow-ing OLT, suggesting the TTR that formsvitreous amyloid may be produced locally.8

Previous reports have established that TTRhas a widespread distribution in the eye butTTR mRNA has exclusively been located inthe retinal pigment epithelium (RPE).9 Giventhat plasma TTR does not cross Bruch’smembrane, it appears that ocular TTR issynthesised at least in part in the RPE, butthe exact factors determining amyloiddeposition are not understood.Bilateral ‘‘branching’’ vitreous deposits of

unknown aetiology should always raise thepossibility of systemic amyloidosis. A relevantfamily history should be sought, but even inits absence, mutations of the TTR gene shouldbe looked for.

H J Zambarakji, D G CharterisMoorfields Eye Hospital, Vitreoretinal Service,

London, UK

Figure 2 Apple-green birefringence withCongo red stain viewed with polarisedmicroscopy of a vitreous biopsy confirming thepresence of amyloid deposits.

Figure 1 Dense nodular white deposits on the posterior capsule of the left eye were presumed tobe of amyloid origin (A), and fundus periphery demonstrates intraretinal haemorrhages andperivascular white deposits (B).

Accepted for publication 1 November 2004

PostScript 773


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

W AyliffeMayday University Hospital, Ophthalmology

Department, London, UK

P J LuthertInstitute of Ophthalmology, Pathology Department,

London, UK

F SchonMayday University Hospital, Neurology Department,

London, UK

P N HawkinsNational Amyloidosis Centre, Department of

Medicine, Royal Free and University College MedicalSchool, London, UK

Correspondence to: Dr H Zambarakji, Angiogenesis,Massachusetts Eye and Ear Infirmary, 325 Cambridge

Street, Boston, MA 02114, USA;[email protected] [email protected]

doi: 10.1136/bjo.2004.057554

References

1 Ando E, Ando Y, Okamura R, et al. Ocularmanifestations of familial amyloidoticpolyneuropathy type I: long term follow up.Br J Ophthalmol 1997;81:295–8.

2 Koga T, Ando E, Hirata A, et al. Vitreousopacities and outcome of vitreous surgery inpatients with familial amyloidotic polyneuropathy.Am J Ophthalmol 2003;135:188–93.

3 Connors LH, Richardson AM, Theberge R, et al.Tabulation of transthyretin (TTR) variants as of 1/1/2000. Amyloid 2000;7:54–69.

4 Benson MD, Turpin JC, Lucotte G, et al. Atransthyretin variant (alanine 71) associated withfamilial amyloidotic polyneuropathy in a Frenchfamily. J Med Genet 1993;30:120–2.

5 Almeida MR, Lopez-Andreu F, Munar-Ques M, etal. Transthyretin ALA 71: a new transthyretinvariant in a Spanish family with familialamyloidotic polyneuropathy. Hum Mutat1993;2:420–1.

6 Kantarjian AD, de Jong RN. Familial primaryamyloidosis with nervous system involvement.Neurology 1953;3:399–409.

7 Rydh A, Suhr O, Hietala SO, et al. Serumamyloid P component scintigraphy in familialamyloid polyneuropathy: regression of visceralamyloid following liver transplantation. Eur J NuclMed 1998;25:709–13.

8 Munar-Ques M, Salva-Ladaria L, Mulet-Perera P,et al. Vitreous amyloidosis after livertransplantation in patients with familial amyloidpolyneuropathy: ocular synthesis of mutanttransthyretin. Amyloid 2000;7:266–9.

9 Cavallaro T, Martone RL, Dwork AJ, et al. Theretinal pigment epithelium is the unique site oftransthyretin synthesis in the rat eye. InvestOphthalmol Vis Sci 1990;31:497–501.

Multifocal electroretinogramdemonstrated macular toxicityassociated with ethambutolrelated optic neuropathyEthambutol is an effective drug in the firstline treatment for tuberculosis but its usemay be associated with ocular toxicity.1 Toxicoptic neuropathy is the most importantocular side effect and is related to the doseand duration of treatment.2 It is usuallybilateral and both central and peripheraltypes of optic neuropathy have beendescribed. The central type involves thepapillomacular bundle and results indecreased visual acuity, caecocentral sco-toma, and blue-yellow colour vision loss,whereas the peripheral type causes peripheralvisual field loss, especially bitemporal defects

with sparing of visual acuity and red-greencolour vision impairment.3 In additional tothe optic nerve toxicity, studies have alsodemonstrated that ethambutol may also betoxic at the retinal level.4–6 We report apatient with ethambutol related toxic opticneuropathy associated with bilateral maculartoxicity as demonstrated by multifocal elec-troretinogram (mfERG). To our knowledge,evaluation of ethambutol related maculartoxicity with mfERG has not been previouslyreported.

Case reportA 45 year old man with pulmonary tubercu-losis presented with a 3 week history ofgradual bilateral visual loss. He has been onantituberculosis therapy with ethambutol900 mg (15 mg/kg/day), rifampicin 600 mg,isoniazid 300 mg, and pyrazinamide 2 g for4 months. The baseline best corrected visualacuity (BCVA) before antituberculosis ther-apy was 20/30 in the right eye and 20/20 inthe left eye. On presentation, the BCVA was20/200 bilaterally without afferent pupillarydefect. Colour vision testing showed redgreen dyschromatopsia. Dilated fundusexamination revealed bilateral small splinterhaemorrhages adjacent to the optic discs withmild left optic disc swelling (fig 1A). Visualfield examination showed bilateral centralscotoma (fig 1B). Magnetic resonance images(MRI) of the brain and orbits were normal. Adiagnosis of toxic optic neuropathy due toethambutol was made.Since the optic discs and fundus changes

were relatively subtle compared with thereduction in visual acuity and the centralvisual field defects and red-green dyschro-matopsia were atypical, further investigations

were performed to evaluate potential asso-ciated retinal toxicity. These included electro-oculogram (EOG), flash electroretinogram(ERG), pattern visual evoked potential(VEP), and mfERG. EOG and flash ERG ofboth eyes were normal. Pattern VEP showedno response for smaller checkers, with weakand delayed response for large checkers of1.8 .̊ MfERG demonstrated bilateral general-ised reduction in N1 and P1 retinal responseamplitudes suggestive of toxic maculopathy(fig 2A).In view of the optic neuropathy and

maculopathy, all antituberculosis drugs werestopped after discussion with the physicians.The patient’s BCVA gradually improved and3 months later, his BCVA improved to 20/30bilaterally. No abnormality was seen onfundus examination. Repeat mfERG record-ing showed recovery of the retinal responsesin both eyes (fig 2B).

CommentThe exact mechanism of ethambutol inducedtoxic optic neuropathy is unclear but it maybe due to retinal ganglion cells or bipolar cellstoxicity at the retinal level.4–7 mfERG is atechnique that allows objective assessment ofmacular function in retinal diseases.8 9 Incontrast with conventional full field flashERG which stimulates and measures theresponse of the entire retina, mfERG stimu-lates individual macular areas and measuresresponses from different retinal locations. Itis useful in differentiating macular and opticnerve diseases as the mfERG in patients withoptic nerve disease should be normal.9 10 Withthe use of mfERG, we demonstrated thatthere was generalised reduction in mfERGresponses at the macula and the areas of


Figure 1 (A) Fundus photograph of the right (left) and left (right) eyes on presentationdemonstrating bilateral splinter haemorrhages (arrows) with mild left optic disc swelling. (B)Automated visual field test showed central scotoma in the left (left) and right (right) eyes.

774 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

abnormality were more extensive than theapparent localised visual field defectsdetected by automated perimetry. The abnor-mal mfERG suggests that in addition to toxicoptic neuropathy, ethambutol may also causemacular toxicity at the neurosensory retinallevel. The improvement in the patient’s BCVAparalleled the improvement in mfERGresponses and therefore mfERG may be auseful tool not only in diagnosis but in theserial assessments of ethambutol relatedocular toxicity, particularly in patients withcentral visual loss.

T Y Y Lai, W-M Chan, D S C LamDepartment of Ophthalmology and Visual Sciences,

The Chinese University of Hong Kong, Hong Kong EyeHospital, Hong Kong, People’s Republic of China

E LimDepartment of Ophthalmology and Visual Sciences,

The Chinese University of Hong Kong, Prince of WalesHospital, Hong Kong, People’s Republic of China

Correspondence to: Dr Wai-Man Chan, Departmentof Ophthalmology and Visual Sciences, The Chinese

University of Hong Kong, 3/F, Hong Kong EyeHospital, 147K Argyle Street, Kowloon, Hong Kong;

[email protected]

doi: 10.1136/bjo.2004.058099

References

1 Carr RE, Henkind P. Ocular manifestations ofethambutol. Toxic amblyopia after administrationof an experimental antituberculous drug. ArchOphthalmol 1962;67:566–71.

2 Leibold JE. The ocular toxicity of ethambutol andits relation to dose. Ann NY Acad Sci1966;135:904–9.

3 Schild H Fox. Raipd-onset reversible oculartoxicity from ethambutol therapy. Am J Med1991;90:404–6.

4 Van Dijk BW, Spekreijse H. Ethambutol changesthe color coding of carp retinal ganglion cellsreversibly. Invest Ophthalmol Vis Sci1983;24:128–33.

5 Kakisu Y, Adachi-Usami E, Mizota A. Patternelectroretinogram and visual evoked corticalpotential in ethambutol optic neuropathy. DocOphthalmol 1987;67:327–34.

6 Nasemann J, Zrenner E, Riedel KG. Recoveryafter severe ethambutol intoxication:psychophysical and electrophysiologicalcorrelations. Doc Ophthalmol1989;71:279–92.

7 Heng JE, Vorwerk CK, Lessell E, et al. Ethambutolis toxic to retinal ganglion cells via an excitotoxicpathway. Invest Ophthalmol Vis Sci1999;40:190–6.

8 Sutter EE, Tran D. The field topography of ERGcomponents in man: I. The photopic luminanceresponse. Vis Res 1992;32:433–46.

9 Hood DC. Assessing retinal function with themultifocal technique. Prog Retin Eye Res2000;19:607–46.

10 Kretschmann U, Bock M, Gockeln R, et al. Clinicalapplications of multifocal electroretinography.Doc Ophthalmol 2000;100:99–113.

Eye involvement mimickingscleritis in a patient with chroniclymphocytic leukaemiaAcute leukaemia is known to affect the eye ina wide variety of ways and detailed post-mortem examination will often reveal sub-clinical involvement. Ocular involvement bychronic leukaemia is much less common.1 Wedescribe the clinical presentation, evaluation,and response to therapy of what is to ourknowledge the first case of clinically signifi-cant scleral infiltration by chronic lymphocy-tic leukaemia mimicking scleritis.

Case reportAn 87 year old man presented with a 2 weekhistory of right eye redness, pain, anddecreased vision. He complained of horizon-tal diplopia for 2 years, which had beencorrected with prisms, but never investigated.Past medical history was notable only forchronic lymphocytic leukaemia (CLL), whichwas first diagnosed 17 years before presenta-tion. He was treated intermittently withchlorambucil, but this was stopped as hiswhite cell count had been stable at 306109/l with minimal cervical and axillarylymphadenopathy. Vision was 6/18 right eyeand 6/9 left eye. Intraocular pressure (IOP)was 18 mmHg in each eye. His vision wasreduced by bilateral early cataracts and hisright eye showed inferior scleral injection(fig 1A), a fibrinous anterior uveitis, andscattered posterior synechiae.The patient was treated with topical

corticosteroids and cycloplegics, but over the3 weeks following presentation his visiondropped to 6/60 right eye because of choroidaleffusions. He had restricted movement of theright eye, proptosis, and shallowing of theanterior chamber with an increase in IOP to26 mmHg. Ultrasound revealed thickening ofthe inferior sclera. A CT scan of the right orbitrevealed a mass in the inferior fornix thatwas attached to the sclera (fig 1B). Excisionbiopsy revealed patchy infiltration by foci ofdensely packed small lymphocytic cells,mainly around small blood vessels (fig 2A).Immunostaining showed a predominance ofweakly stained CD20 positive B lymphocyteswith scattered CD3 positive T cells consistentwith CLL. Haematological analysis revealed atotal white cell count of 49.16109/l. Hereceived orbital radiotherapy 27.5 Gy in 11fractions using 6 MV x rays, completing thecourse 5 months after presentation. Theinflammation, choroidal effusions, and orbi-tal mass regressed completely but he was leftwith thinning of the inferior sclera (fig 2B).He was last seen 2 years after presentationwith vision of 6/60 right eye but no evidenceof active ocular involvement from his CLL. Herefused cataract surgery.

Figure 2 (A) Trace arrays of multifocal electroretinogram (mfERG) of the right (left) and left (right)eyes on presentation showing generalised reduction of retinal responses at the macula. (B) Tracearrays of mfERG of the right (left) and left (right) eyes 3 months after cessation of ethambutoldemonstrating recovery and improvement of the mfERG responses.


Financial support: nil.

Financial interest: nil.

Figure 1 (A) Right eye showing diffuseinjection of the inferior sclera. (B) Computedtomography (CT) scan of orbit showing a massthat was confluent with the sclera.

PostScript 775


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

CommentThis patient presented with what at firstappeared to be a scleritis with a secondaryfibrinous anterior uveitis. He was 87 yearsold, which is an unusual age for a firstpresentation of either scleritis or anterioruveitis. His history of chronic diplopia andpast medical history of CLL suggested apossible aetiology for these findings. It isuncommon for CLL to affect the eye in aclinically significant way2 although postmor-tem examination has shown relatively fre-quent subclinical involvement in between30% and 90% of patients.3 Despite this, CLLhas been described causing uveitis, glaucoma,iris infiltration,4 retinal detachments, con-junctival vascular changes, optic neuropathy,and orbital involvement as well as the moretypical retinal haemorrhagic changes.Episcleral and scleral infiltration is recog-

nised in acute leukaemia and has been seenat necropsy in chronic leukaemia in 14% ofcases in one series but this is rarely sympto-matic.3

In our patient the response to radiotherapywas encouraging and although his finalacuity was not improved this was due in partto the progression of his cataracts. Insummary, this is the first case we are awareof in which CLL had infiltrated the sclera ofan eye to a sufficient degree that it simulatedscleritis. The scleral mass regressed comple-tely following radiotherapy although visiondid not recover.

B J L BurtonMedical Retina Service, Moorfields Eye Hospital, City

Road, London EC1V 2PD, UK

E T Cunningham JrDepartment of Ophthalmology, New York University

School of Medicine, New York, NY, USA

I A CreeInstitute of Ophthalmology 11–43 Bath Street, London

EC1V 9EL, UK

C E PavesioMedical Retina Service, Moorfields Eye Hospital, City

Road London EC1V 2PD, UK

Correspondence to: Mr C Pavesio, Moorfields EyeHospital, City Road, London EC1V 2PD, UK;

[email protected]

doi: 10.1136/bjo.2004.060152

References

1 Allen RA, Straatsma BR. Ocular involvement inleukemia and allied disorders. Arch Ophthalmol1961;66:490–508.

2 Buchan J, McKibbin M, Burton T. The prevalenceof ocular disease in chronic lymphocyticleukaemia. Eye 2003;17:27–30.

3 Kincaid MC, Green WR. Ocular and orbitalinvolvement in leukemia. Surv Ophthalmol1983;27:211–32.

4 Martin B. Infiltration of the iris in chroniclymphatic leukaemia. Br J Ophthalmol1968;52:781–5.

Regression of choroidalmetastases from breastcarcinoma using aromataseinhibitorsBreast cancer has become a leading healthconcern in the United States, accounting for30% of all cancers among women.1 Oestrogenis an important hormone involved in thedevelopment and growth of breast tumours.2

It has been found that 60–70% of breastcancers have oestrogen receptors.2 Followingtumour resection, receptor positive patientsare commonly treated with hormone therapysuch as tamoxifen. Tamoxifen is a competi-tive antagonist of oestrogen at its receptorsite. It is most often used for postmenopausalpatients with oestrogen receptor positivebreast cancers. However, tamoxifen also hasa partial oestrogen agonist effect which couldbe detrimental, because it can lead toincreased risk for uterine cancer, throm-boembolism, and treatment failure.3

The third generation aromatase inhibitors,including anostrozole, letrozole, and vorozole,have emerged as a new treatment for post-menopausal women with oestrogen receptorpositive breast cancer. Their mechanismdiffers from that of tamoxifen as theyminimise peripheral conversion of circulatingandrogen to oestrogen.3 We report onepatient with choroidal metastasis from oes-trogen receptor positive breast cancer whoshowed an excellent response to oral aroma-tase inhibitors.

Case reportA 66 year old woman presented in January2003 with a 5 month history of flashinglights in the right eye. Her medical historyrevealed breast cancer, oestrogen receptorpositive, treated with modified radical mas-tectomy and tamoxifen. In 1989, tamoxifenwas stopped per the standard 5 year protocol.In October 2002, she developed metastases tothe supraclavicular lymph nodes bone andwas started on anastrozole.On examination, visual acuity was 20/60

RE and 20/20 LE. Fundus examinationrevealed subretinal fluid overlying a solitaryamelanotic choroidal metastasis measuring12 mm in base and 3 mm in thickness (fig 1).The patient was continued on anastrozole. In

September 2003, the choroidal metastasiscompletely regressed with resolution of sub-retinal fluid (fig 2). Her visual acuityremained 20/40 RE and 20/20 LE.

CommentUveal metastases are the most commonintraocular malignancy. They typically affectthe posterior choroids.4 The most commonprimary sites of cancer are from breast (47%),lung (21%), and gastrointestinal tract (4%).5

Classically, choroidal metastases are yellow,plateau shaped, with secondary subretinalfluid. The treatment of choroidal metastases

Figure 2 September 2003. (A) Following9 months of anastrozole treatment, choroidalmetastasis has regressed.(B) Optical coherencetomography following 9 months of anastrozoletreatment showing resolution of subretinal fluid.


Figure 2 (A) Histopathology of scleral massshowing foci of densely packed smalllymphocytic cells, mainly around small bloodvessels (haematoxylin and eosin stain, originalmagnification6200). (B) Right eye followingtherapy showing thinning of the inferior sclera.

Figure 1 January 2003. (A) Active choroidalmetastasis from breast carcinoma at firstexamination. (B) Optical coherencetomography showing subretinal fluid in thefovea.

776 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

depends on many factors including location,multiplicity, and activity of each tumour.6 7

Additionally, the projected visual outcomeand underlying systemic control is important.In some cases, therapy is limited to the eye,especially if systemic metastases are absent orin remission. In these instances, externalbeam radiotherapy or plaque radiotherapy areemployed.6 In most instances, however, thepatient has systemically active metastaticdisease so therapy is directed towards treat-ment of both the ocular and systemic diseaseusing chemotherapy or hormone therapy.6

The most commonly employed hormonaltreatment for breast cancer is tamoxifen.2 It isusually employed for postmenopausalpatients with breast cancer who displayoestrogen receptors. The effect of tamoxifenis due to its anti-oestrogenic activity, bycompetitive inhibition of oestrogen bindingto oestrogen receptors.2 Tamoxifen inhibitsthe expression of oestrogen regulated genesincluding growth factors and angiogenicfactors secreted by the tumour. Tamoxifencan also induce programmed cell death.Tamoxifen additionally has partial agonisteffects, which could be beneficial because itprevents bone demineralisation in postmeno-pausal women.3 However, these oestrogeniceffects are also associated with increasedrisks of uterine cancer, thromboembolism,and treatment failure.3

Tamoxifen is well tolerated by mostpatients with breast cancer and only 5% ofpatients note related menopausal symptoms,such as hot flashes and vaginal discharge.2

Retinopathy has been reported in womenwith high doses of tamoxifen, but not withconventional doses.In postmenopausal women the main

source of oestrogen is from peripheral con-version of adrenal androgen. New thirdgeneration aromatase inhibitors, includinganastrozole (Arimidex, Zeneca) and letrozole(Femara, Novartis), act by preventing thisconversion, thus lowering circulating oestro-gen levels. Studies have shown that they areequal or superior to tamoxifen in clinicallyefficacy for metastatic breast carcinoma.3

They are well tolerated and have been shownto have important benefits over tamoxifen.Comparative trials indicate that anastrozolehas similar adverse effects compared totamoxifen.8 9 Our patient responded drama-tically to aromatase inhibitors after failing torespond to tamoxifen. Based on this case andthree other cases with a favourable responsewe anticipate that aromatase inhibitors couldbe a promising alternative for patients withchoroidal metastases from oestrogen receptorpositive breast carcinoma.

M E Manquez, C L Shields, E C Karatza,J A Shields

Oncology Service, Wills Eye Hospital, ThomasJefferson University, Philadelphia, PA, USA

Correspondence to: Dr Maria Manquez, OncologyService, Wills Eye Hospital, 840 Walnut Street,

Philadelphia, PA 19107, USA;[email protected]

doi: 10.1136/bjo.2004.061127

References

1 Feig BW, Berger DH, Fuhrman GM. In: Breastcancer. The MD Anderson surgical oncologyhandbook. Philadelphia: Lippincott Williams andWilkins, 2003:14.

2 Osborne CK. Tamoxifen in the treatment of breastcancer. N Engl J Med 1998;339:1609–18.

3 Smith IE, Dowsett M. Aromatase inhibitors inbreast cancer. N Engl J Med 2003;348:2431–42.

4 Shields JA, Shields CL. In: Metastatic tumor to theuvea and retina. Intraocular tumors. A text andatlas. Philadelphia: WB Saunders,1992:278–320.

5 Shields CL, Shields JA, Gross N, et al. Survey of520 eyes with uveal metastases. Ophthalmology1997;104:1265–76.

6 Shields CL. Plaque radiotherapy for themanagement of uveal metastasis. Curr OpinOphthalmol 1998;9:31–7.

7 Demirci H, Shields CL, Chao AN, et al. Uvealmetastasis from breast cancer in 264 patients.Ophthalmology 2003;136:264–71.

8 ATAC (Arimidex, Tamoxifen Alone or inCombination) Trialists’ Group. Anastrozole aloneor in combination with tamoxifen alone foradjuvant treatment of postmenopausal womenwith early breast cancer: first results of the ATACrandomized trial. Lancet 2002;359:2131–9.

9 Mouridsen H, Gershanovich M, Sun Y, et al.Superior efficacy of letrozole versus tamoxifen asfirst-line therapy for postmenopausal women withadvanced breast cancer: results of a phase IIIstudy of the International Letrozole Breast CancerGroup. J Clin Oncol 2001;19:2596–606.

‘‘Fingertip’’ cryoprobe assistedorbital tumour extractionCryoprobes are used to grasp, provide trac-tion, and facilitate orbital tumour removal.1–4

Standard ophthalmic cryoprobe tips aretypically rounded and offer small surfaceareas for cryo-adhesion. I describe a newspatulated cryotherapy probe with a largeoval and uniform surface area for adhesion

and subsequent traction. This report usesphotography to document the use of thisprobe and describes its manufacture andeffectiveness during orbital surgery.

MethodsThree spatulated probes were manufactured(Mira, Inc, ‘‘Finger-tip probe,’’ Uxbridge, MA,USA) under Good Manufacturing Practice(GMP), International Standards Organiza-tion (ISO), and Food and Drug Administra-tion (FDA) guidelines (fig 1).The medium sized probe has an active

surface of 25.2 mm2. Its design allows forpreferential cooling of the active surface ofthe applicator. Mira tested this probe to coolto up to –85 C̊ on nitrous oxide (as governedby the Joule-Thompson principle).

ResultsA 77 year old patient was found to have asuperonasal orbital tumour (by clinical exam-ination and radiographic imaging). Thetumour was 18 mm in largest diameter andfound to be contiguous with both the orbital

the Macula Society (Dr CL Shields), the Paul KayserInternational Award of Merit in Retina Research,Houston, TX (Dr JA Shields).

Figure 1 Top, the 25.2 mm2 surface arealarge ‘‘Finger-tip’’ cryotherapy probe set toshow its spatulated active surface. Bottom,computed tomography shows a well defined,1.8 cm lobular tumour in the superonasal orbit.

Figure 2 Top left, the tumour (arrow) can be seen deep within the superonasal orbit. Top right, thecryoprobe is introduced such that the spatulated tip is placed flush with the tumour and not touchingadjacent orbital tissues. Bottom left, the tumour is delivered from the orbit utilising the cryo-adhesion. Bottom right; the tumour is visualised outside the orbit. Note the freeze within the tissuewith minimal ice formation on the shaft of the probe.


Supported in part by the Pan American Ophthalmo-logy Foundation and Retina Research Foundation (DrME Manquez), the Eye Tumor Research Foundation,Philadelphia, PA (Dr CL Shields) the Macula Foundation,New York, NY (Dr CL Shields), the Rosenthal Award of

PostScript 777


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

portion of the frontal bone and optic nerve.Ultrasonography revealed intrinsic tumourvascularity (low flow). Cavernous haeman-gioma was suspected and an anterior orbi-totomy performed. A combination of sharpand blunt dissection was required to exposethe tumour. Then, a medium sized ‘‘Finger-tip’’ cryoprobe was used to create an adhe-sion to the tumour’s surface (fig 2). Thisrelatively large cryo-adhesion allowed foreasier handling of the tumour and improvedtraction (fig 2). The tumour remained adher-ent to the cryoprobe for 60 seconds after thenitrous oxide was discontinued. Freezinginduced no change in the appearance ofthe tumour on gross examination.Histopathology revealed cavernous haeman-gioma. No clinical side effects (related tocryo-extraction) were noted.

CommentCryo-extraction works best with tumours andcysts that contain fluids (for example, blood,tears), as opposed to solid tumours and thosecontaining lipid or keratin (for example,dermoid). Thus, cavernous haemangiomasare particularly good candidates becausefreezing occurs on both the tumour’s surface(capsule) and within the tumour stroma.With a well formed fibrous capsule, the outerand inner ice balls provide a strong attach-ment to the probe.The fingertip cryoprobes differ from pre-

viously available applicators in that they offerrelatively large, spatulated tips. This allowsfor more homogeneous cryo-adhesion over alarger surface area. The larger surface area ofapplication also increases tissue penetration,resulting in a stronger adhesion.This study examines a new type of cryop-

robe used to assist orbital surgery duringtumour extraction. It was found capable ofadhering to the tumour. Held by the surgeon,tumour traction could be induced with onehand, while the other hand wielded instru-ments used to sever residual tumour attach-ments.5 Fingertip cryoprobes offer analternative to current devices used for cryo-extraction of orbital tumours.

Correspondence to: P T Finger, MD, The New YorkEye Cancer Center, 115 East 61st Street, New York

City, NY 10021, USA; [email protected]

doi: 10.1136/bjo.2004.055426

References

1 Maroon JC, Onik G, Quigley MR, et al.Cryosurgery re-visited for the removal anddestruction of brain, spinal and orbital tumours.Neurol Res 1992;14:294–302.

2 Henderson JW, Neault RW. The use of thecryoprobe in the removal of posterior orbitaltumors. Ophthalmic Surg 1976;7:45–7.

3 Loewenstein A, Geyer O, Lazar M. Cavernoushaemangioma of the orbit: treatment bytransconjunctival cryoextraction. Eye 1993;7(Pt4):597–8.

4 Gdal-On M, Gelfand YA. Surgical outcome oftransconjunctival cryosurgical extraction of orbitalcavernous hemangioma. Ophthalmic Surg Lasers1998;29:969–73.

5 Finger PT. ‘‘Finger-tip’’ cryoprobe assistedenucleation. Am J Ophthalmol, (in press).

Congenital third nerve palsy,moyamoya disease and opticnerve head staphylomaThe association of congenital optic nervehead anomalies, especially of the morningglory disc variety, with moyamoya disease iswell recognised and has been described in anumber of patients.1–4 To the best of ourknowledge, the occurrence of a congenitalthird nerve palsy with moyamoya disease anda congenital optic nerve head anomaly hasnot been reported. We describe a patient whodemonstrated the ipsilateral occurrence ofthese three congenital abnormalities,strengthening the association of congenitaloptic nerve and carotid abnormalities andsuggesting a common underlying aetiology.

Case reportA 3 year old boy was examined because ofleft sided exotropia and blepharoptosis. Hisbirth, developmental, and medical historieswere unremarkable. Visual acuity was 20/25right eye and 20/200 left eye. There was anear total ptosis of the left upper eyelid. Therewas a large exotropia and a small hypotropiaof the left eye that he was unable to adduct,elevate, or infraduct. His right pupil reactedbriskly to light, but the left pupil was dilatedand sluggishly reactive to light. A left relativeafferent pupillary defect was present. Fundusexamination disclosed an anomalous leftoptic disc and peripapillary area, with whatappeared to be an enlarged scleral openingand a staphylomatous defect around the opticpapilla (fig 1). The rest of the retina wasnormal. The right fundus was normal.A magnetic resonance imaging scan and

magnetic resonance angiogram of the brainrevealed an absence of the left intracranialcarotid artery and its bifurcation into middleand anterior cerebral arteries. The lenticulo-striate arteries were increased in size, con-sistent with moyamoya vessels (fig 2).

CommentThis patient has an optic nerve head mal-formation that appears to be most compatiblewith a peripapillary staphyloma or a variantof a morning glory disc anomaly (MGDA).MGDA is one of the cavitary optic discmalformations comprising a congenital, fun-nel-shaped excavation of the posterior peri-papillary sclera that incorporates the opticdisc. In addition to the anomalous retinal

vasculature that distinguishes the MGDAfrom other excavated optic disc anomaliessuch as optic disc coloboma, the associationof morning glory disc with carotid circulationanomalies, especially moyamoya disease hasbeen reported.1–5 Moyamoya disease is a rarecerebrovascular disorder characterised bystenosis or occlusion of the distal internalcarotid arteries. Progressive brain ischaemiatriggers formation of a collateral vascularnetwork in the basal ganglia region referredto as moyamoya (Japanese word for ‘‘cloud ofsmoke’’) vessels. The most common clinicalfeatures are transient ischaemic attacks andstroke in children, and intracranial haemor-rhage in adults. Our patient appears to havean extreme form of this condition with totalabsence of the left internal carotid artery andcollateral vessel formation typical of moya-moya disease.Congenital third nerve palsy was once

thought not to be associated with otherneurological abnormalities. More recent stu-dies have however revealed a high incidenceof associated neurological deficits such ashemiparesis, seizures, hemianopia, andhydrocephalus.5 An association of congenitalthird nerve palsy with MGDA or peripapillarystaphyloma has not been reported to the bestof our knowledge. The third nerve palsy inthe present case may be due to an ischaemicbrainstem event or a developmental defectaffecting the left third nerve nucleus.The constellation of abnormalities in our

patient confirms the association of moya-moya vessels with optic nerve head malfor-mations such as MGDA or peripapillarystaphyloma, and the possible neurologicalcomplications of this malformation complex,as evidenced by the third nerve palsy in thiscase. It is possible that an intracranialvascular dysgenesis may underlie some casesof the morning glory disc anomaly. Westrongly recommend magnetic resonanceangiography in conjunction with magneticresonance imaging to identify carotid vascu-lar anomalies in patients with MGDA orperipapillary staphylomas, especially in thepresence of other neurological signs such as acongenital third nerve palsy.

K Sabti, B A HajjAl-Bahar Eye Center and the Department of

Ophthalmology, University of Kuwait, Kuwait City,Kuwait

J-M HwangDepartment of Ophthalmology, Seoul MunicipalBoramae Hospital, College of Medicine Seoul

National University, Seoul, Korea


The work is supported by The EyeCare Foundation, Incand Research to Prevent. Blindness, New York, USA.

Dr Finger has no proprietary interest in the instrumentdescribed in this study.

Figure 2 Cerebral angiogram shows totalabsence of left internal carotid artery (arrow)with increased size of lenticulostriate arteriescompatible with moyamoya disease.

Figure 1 Enlarged optic nerve head scleralopening with radial exit of retinal blood vessels.Note staphylomatous appearance ofperipapillary area.

778 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

J-M Hwang, E I TraboulsiThe Center for Genetic Eye Diseases, Cole Eye

Institute, Cleveland, OH, USA

J ReidDepartment of Pediatric Radiology, The Cleveland

Clinic Foundation, Cleveland, OH, USA

Correspondence to: Elias I Traboulsi, MD, Center forGenetic Eye Disease, Cole Eye Institute, The Cleveland

Clinic Foundation, i32, 9500 Euclid Avenue,Cleveland, OH 44195, USA; [email protected]

doi: 10.1136/bjo.2004.059246

References

1 Hanson MR, Price RL, Rothner AD, et al.Developmental anomalies of the optic disc andcarotid circulation. A new association. J ClinNeuroophthalmol 1985;5:3–8.

2 Massaro M, Thorarensen O, Liu GT, et al.Morning glory disc anomaly and moyamoyavessels. Arch Ophthalmol 1998;116:253–4.

3 Bakri SJ, Siker D, Masaryk T, et al. Ocularmalformations, moyamoya disease, and midlinecranial defects: a distinct syndrome.Am J Ophthalmol 1998;116:253–4.

4 Krishnan C, Roy A, Traboulsi E. Morning glorydisk anomaly, choroidal coloboma, andcongenital constrictive malformations of theinternal carotid arteries (moyamoya disease).Ophthalmic Genet 2000;21:21–4.

5 Tsaloumas MD, Willshaw HE. Congenitaloculomotor palsy: associated neurological andophthalmological findings. Eye 1997;11(Pt4):500–3.

Radial optic neurotomy for thetreatment of acute functionalimpairment associated with opticnerve drusenRadial optic neurotomy was recently intro-duced as a treatment option in patients withcentral retinal vein occlusion.1 As describedby Opremcak et al,1 central retinal veinocclusion might be related to increasedpressure on the central retinal artery andvein as well as on optic nerve fibres in theconfined space provided by the scleral ring. Itwas therefore suggested that a relaxation ofthe scleral outlet by a radial optic neurotomymight be an effective surgical treatmentoption. In the light of this information we

hypothesised, that radial optic neurotomymay also be applicable in patients with visualfield defects and deterioration of visual acuityassociated with optic nerve drusen, wherecompression induced damage to optic nervefibres is the underlying pathogenetic princi-ple2

Case reportA 27 year old female patient presented withextensive bilateral optic nerve drusen (fig 1A).While visual acuity was light perception onthe right eye over 4 years, she had experi-enced an acute and rapid deterioration ofvisual acuity from 20/32 to 20/500 and aprogressive visual field loss within the past6 weeks before she was seen in our institu-tion (fig 1B). The progressive visual fielddefect had been documented carefully by thereferring ophthalmologist. Besides the opticnerve drusen, there was no other ophthalmicpathology, no history of glaucoma, or anyother relevant disease. With respect to therapid functional deterioration we discussedradial optic neurotomy as a potential treat-ment option. Surgery was performed, afterwritten informed consent, by one of theauthors (AK) and consisted of standard threeport pars plana vitrectomy and radial opticneurotomy. The incision was performed atthe nasal edge of the optic disc in a radialfashion avoiding major retinal vessels. Therewere no intraoperative complications except asmall haemorrhage at the incision site. Thepatient was then carefully followed post-operatively at 3 week intervals. Already at thefirst follow up visit a regression of the visualfield defect and slight improvement of visualacuity was noted (fig 2A). At last presenta-tion 10 weeks after surgery, the visual fielddefect was limited to the inferior nasalquadrant and visual acuity had improved to20/32 (fig 2B). Funduscopy revealed a scar atthe incision site, but no other pathologies(fig 2C).

CommentDrusen of the optic nerve are the result ofaxonal degeneration of retinal ganglioncells and are composed of acellular concentriccalcified laminations.3 While most cases areidiopathic, drusen have been described inassociation with several acquired condi-tions such as hypertensive retinopathy, vas-cular occlusion, optic atrophy, or chronic

papilloedema.4 If drusen are located deep inthe optic nerve, progressive enlargement mayresult in pressure induced atrophy of adjacentnerve fibres producing visual field constric-tion and anterior ischaemic optic neuropa-thy.2 Drusen have also been reported to occurpredominantly in eyes with abnormally smalloptic discs.5 It had been previously hypothe-sised,1 that a radial incision at the nasal edgeof the optic disc might result in a decom-pression of the scleral outlet and the asso-ciated neurovascular compression in patientswith central retinal vein occlusion.Transferring this hypothesis to our patient,radial optic neurotomy seemed a reasonabletherapeutic approach to us, as a relaxation ordecompression of the scleral outlet mightcounteract the relevant pathogenetic princi-ples of progressive visual field defect in opticnerve drusen. To our knowledge, radial opticneurotomy has not been used for the treat-ment of progressive visual field defects inassociation with optic nerve drusen before.We were very aware of the experimentalcharacter of this surgical intervention andcarefully informed the patient before theoperation. However, with respect to the rapiddeterioration of visual acuity and progressionof visual field constriction, there seemed to beno other treatment option available and thesurgical intervention appeared justified.The validity of the ‘‘scleral outlet compart-

ment syndrome’’ concept in central retinalvein occlusion, as well as the effect of theradial neurotomy, has been questioned6 byHayreh and is currently under discussion.However, the excellent functional outcome inthis case may suggest that radial opticneurotomy potentially provides a relaxationand decompression of the optic nerve, allow-ing a recovery in conditions associated withpressure induced nerve fibre damage as in thecase presented. In retinal vein occlusion,other additional mechanisms of action, suchas the formation of chorioretinal shunts as aresult of radial optic neurotomy, weredescribed to contribute to visual recovery.7

In summary, this case reports indicatesthat radial optic neurotomy might be con-sidered in patients presenting with suddenvisual loss and constriction of visual field inassociation with optic nerve drusen. Weconsidered the surgical approach in thissingle case, as there was no other treatmentoption we could offer the patient that mightrestore vision.

10

0

–10

–20

–30

100%

30°

LE

BA

dB

0 20 40 60 80

Figure 1 Highly reflective signal seen during B-scan echography at the optic nerve head of the left eye as typical for optic nerve drusen (A). The drusenwere associated with a concentric visual field defect (B), visual acuity was 20/500.


PostScript 779


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

C Haritoglou, S G Prieglinger, M Grueterich,A Kampik

Department of Ophthalmology, Ludwig-Maximilians-University, Mathildenstrasse 8, 80336 Munich,

Germany

G K KriegelsteinDepartment of Ophthalmology, University of Cologne,

Joseph-Stelzmann-Strasse 9, 50931 Cologne,Germany

Correspondence to: Christos Haritoglou, MD,Department of Ophthalmology,

Ludwig-Maximilians-University, Mathildenstrasse 8,80336 Munich, Germany;

[email protected]

doi: 10.1136/bjo.2004.060335

References

1 Opremcak EM, Bruce RA, Lomeo MD, et al.Radial optic neurotomy for central retinal veinocclusion. A retrospective pilot study of 11consecutive cases. Retina 2001;21:408–15.

2 Moody TA, Irvine AR, Cahn PH, et al. Suddenvisual field constriction associated with optic discdrusen. J Clin Neuroophthalmol 1993;13:8–13.

3 Tso MOM. Pathology and pathogenesis of drusenof the optic nerve head. Ophthalmology1981;88:1066–80.

4 Apple DJ, Rabb MF, Walsh PM. Congenitalanomalies of the optic disc. Surv Ophthalmol1982;27:3–41.

5 Jonas JB, Gusek G, Guggenmoos-Holzmann I,et al. Optic nerve head drusen associated withabnormally small optic discs. Int Ophthalmol1987;11:79–82.

6 Hayreh SS. Radial optic neurotomy for centralretinal vein occlusion. Retina 2002;22:374–7.

7 Garcia-Arumi J, Boixadera A, Martinez-Castillo V, et al. Chorioretinal anastomosis afterradial optic neurotomy for central retinal veinocclusion. Arch Ophthalmol 2003;121:1385–91.

Early chorioretinal anastomosisin non-ischaemic CRVO: arandomised trialIn non-ischaemic central retinal vein occlu-sion (niCRVO), the two principal determi-nants of final visual acuity are visual acuityand the presence of macular oedema at initialpresentation.1 2 Data from the Central VeinOcclusion Study Group2 suggested that ofpatients with CRVO with an initial acuitybetter than 6/15, 65% maintain this, whereaspatients with presenting acuities between6/15 and 6/60, 81% remain the same or getworse (19% improve to 6/12 or better, 44%remain between 6/15, and 6/60 and 37% areworse than 6/60). Up to 34% of niCRVOmay progress to the ischaemic variantwith its attendant complications within3 years.2 However, in some studies 83% ofindeterminate cases progress to ischaemicCRVO.1 The creation of a chorioretinal venousanastomosis (CRVA) improved visual acuityin some patients, decreased macular oedema,

and reduced the incidence of progressionto ischaemic CRVO.3–5 These importantobservations provide the basis for the studywe carried out.During the study period 11 patients

(table 1) were enrolled according to the trialprotocol (see appendix). All anastomoseswere patent (on fluorescein angiographyand if not repeated, see appendix). Meanpatient age and mean pretreatment (T0)visual acuities, retinal thickness (by opticalcoherence tomography, OCT), and cyst height(OCT) did not differ between the two groups(table 1, fig 1). Factorial analysis of variance(ANOVA, Genstat) found that all threemeasures (acuity, retinal thickness, and cystheight, fig 1) decreases significantly betweenT0 and 6 months, but only in the case of cystheight was there a significant differencebetween the laser treated and control groups(F(1,9)=5.85, p,0.05). The changes inretinal thickness and visual acuity were alsogreater in the treatment group but did notachieve significance (F(1,9)=2.51 and 1.15,respectively, both p.0.1). Larger group sizesmay have shown a significant effect.The principal limitation of this trial is the

small number of patients, because of thelimited numbers of suitable patients withinthe study period. However, patients wererandomised and well matched. The trialshowed a trend towards better vision over6 months in the laser treated group, althoughthis failed to reach statistical significance.


Financial interest: none.

10

0

–10

–20

–30

100%

30°

LE

C

A

B

dB

0 20 40 60 80

10

0

–10

–20

–30

100%

30°

LEdB

0 20 40 60 80

Figure 2 There was a regression of the visual field defect noted 3 weeks after radial optic neurotomy (A). Ten weeks postoperatively the visual fielddefect was predominantly limited to the lower nasal quadrant (B). Funduscopy revealed a scar at the incision site at the nasal rim of the optic disc (C).

780 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

It remains possible that a larger groupof randomised patients would have pro-duced statistically significant numbers. Amulticentre trial would be required to achievethis. Another question arises regarding thelonger term outcome in these patients. Againthis should probably be considered in furtherstudies. Cyst height might predict later visualacuity as there is evidence that it correlateswith visual acuity outcome measures.6 7 It istherefore possible that longer follow up mayhave found a significant improvement inacuity. Retinal thickness predicts and pre-cedes changes in visual acuity in diabeticretinopathy.8

If the size and duration of this trial hadbeen increased, it is conceivable that resultsmay have proved more informative. The factthat there was a significant effect on cystheight on OCT is, nevertheless, importantand suggests that laser CRVA may reducemacular oedema in CRVO. The finding thatimprovements in visual acuity are recordedlater than reduction in macular oedema8 inother studies implies that a significant effecton visual acuity might have been expected inthese patients at a later time. We thereforerecommend that this be taken into account infuture studies.

AcknowledgementsWe gratefully acknowledge the help of ProfessorStephen Dunnett (Cardiff University) with statis-tical analysis of the data.

R J Antcliff, E J Mayer, T H Williamson,J S Shilling

Department of Ophthalmology, St Thomas’s Hospital,Lambeth Palace Road, London E1 7EH, UK

Correspondence to: Eric Mayer, University of Bristol,Clinical Sciences, Bristol Eye Hospital, Lower Maudlin

Street, Bristol BS1 2LX, UK; [email protected]

doi: 10.1136/bjo.2004.062539

AppendixThe principal study end point was visualacuity at 6 months and resolution of macularoedema, the secondary end point was theidentification of subgroups on optical coher-ence tomography (OCT 2000, HumphreyInstruments, USA) with a more favourableoutcome including reduction of consecu-tive ischaemia. For entry into the studypatients were required to: (1) be within1 month of symptom onset; (2) have visualacuity of 6/24 or worse; (3) have good retinalperfusion (less than 10 disc diameters ofnon-perfusion); (4) have no relative afferentpupillary defect or neovascularisation; (5)have macular oedema confirmed by opticalcoherence tomography and fluorescein

fundus angiography (FFA). The followingexclusion criteria were applied: diabetes withmacular oedema in the fellow eye; inability togive informed consent; pregnancy; sensitivityto fluorescein; cloudy media sufficient topreclude adequate fundal photography; ageless than 40 years.Upon entry into the trial the following

were performed: best corrected visual acuity(Snellen and ETDRS); fundus photography,and fluorescein angiography; OCT assess-ment of the macula. Patients were thenrandomised (envelope) to laser or observa-tion groups. Laser anastomosis was per-formed with a 50 mm spot size for0.1 second with 1.5–2.5 W (maximum powerpossible) with an argon laser. The firstshot was fired at the edge of the vein torupture Bruch’s membrane and the secondshot at the edge of the vein to rupture thevein. A third shot with a YAG laser wasapplied (3–5 mJ) if required. Anastomoticsites were attempted in the following loca-tions in order of preference—inferonasal,inferotemporal, or superonasal. All anasto-moses were made at least 3 disc dia-meters from the optic disc. At 3–4 weeksrepeat FFA was performed to determineanastomotic success: where no anastomosiswas present this was repeated, until func-tioning.Patients were reassessed at 1, 2, 3, 6, and

12 months and acuities and OCT recorded.It was expected that approximately 40

patients would be required to show a 50%benefit.Because of laser surgery it was not possible

to mask the patient or observer as to whichtreatment had been carried out.No adverse events were reported or noted

in this series.

References

1 The Central Vein Occlusion Study. Baseline andearly natural history report. Arch Ophthalmol1993;111:1087–95.

2 The Central Vein Occlusion Study Group. Naturalhistory and clinical management of central retinalvein occlusion. Arch Ophthalmol1997;115:486–91.

3 McAllister IL, Constable IJ. Laser-inducedchorioretinal venous anastomosis fortreatment of nonischemic central retinalvein occlusion. Arch Ophthalmol1995;113:456–62.

4 McAllister IL, Douglas JP, Constable IJ, et al.Laser-induced chorioretinal venous anastomosisfor nonischemic central retinal vein occlusion:evaluation of the complications and theirrisk factors. Am J Ophthalmol 1998;126:219–29.

5 McAllister IL, Vijayasekaran S, Yu DY, et al.Chorioretinal venous anastomoses: effect ofdifferent laser methods and energy in humaneyes without vein occlusion. GraefesArch Clin Exp Ophthalmol 1998;236:174–81.

6 Antcliff RJ, Spalton DJ, Stanford MR, et al.Intravitreal triamcinolone for uveitic cystoidmacular edema: an optical coherencetomography study. Ophthalmology2001;108:765–72.

7 Beausencourt E, Remky A, Elsner AE, et al.Infrared scanning laser tomography ofmacular cysts. Ophthalmology2000;107:375–85.

8 Terasaki H, Kojima T, Niwa H, et al. Changes infocal macular electroretinograms and fovealthickness after vitrectomy for diabetic macularedema. Invest Ophthalmol Vis Sci2003;44:4465–72.

ObservationLaser

1000

800

600

400

200

0

A

T0 T6 months

500

400

300

200

100

0

B

T0 T6 months

1.0

0.8

0.6

0.4

0.2

0.0

C

T0 T6 months

Figure 1 Data of retinal thickness (A), cyst height (B), and visual acuity (C) for patients randomisedto laser chorioretinal venous anastomosis or observation. Visual acuity is measured on the ETDRSscale. Retinal thickness and cyst height were measured in pixels (1 pixel = 4 mm). Error bars showstandard errors of means, obtained from factorial analysis of variance. Significant effects areidentified as follows: *p,0.05; **p,0.01.

Table 1 Comparison of randomised groups for early laser chorioretinal venousanastomosis versus observation in non-ischaemic CRVO (patient data areillustrated, visual acuities are given as ETDRS measures)

Parameter Control Laser t test

Number of patients 5 6Female patients 2 2Left eyes 5 3Mean age 76.4 68.8 1.278, p = 0.117, NSMean pretreatment acuity 0.872 0.827 0.216, p = 0.417, NSMean 6 month acuity 0.664 0.41 1.388, p = 0.099, NSMean change in acuity over 6 months 0.208 0.417 21.074, p = 0.155, NSMean change in retinal thickness onOCT over 6 months

172.8 459.3. 21.583, p = 0.074, NS

Mean change in cyst height onOCT over 6 months

87.6 302.6. 22.418, p = 0.019*

OCT, optical coherence tomography.

Accepted for publication 14 January 2005

RJA was supported by the Lady Allerton Fund.

Competing or financial interests: none.

PostScript 781


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

Informed consent and medicaldevices: the case of the contactlensIt is estimated that 1.65 million people in theUnited Kingdom wear contact lenses, 97% ofwhom do so for refractive and cosmeticreasons.1 Contact lens wearers are at risk ofsight threatening adverse events such ascorneal ulceration, which may lead to scar-ring and visual loss. The risk of cornealulceration is approximately 1 in 3000, 1 in2000, and 1 in 500 patients per year for hard,soft daily wear, and soft continuous wearlenses, respectively.2–4

Informing individuals of the risks asso-ciated with contact lens wear, is importantboth for the wearer and person prescribingthe contact lens. For consent to be legallyvalid it must be informed. This means thatpatients need information about the natureof an intervention, possible alternatives and,specifically, its benefits and risks. There is noreason why the principle of informed consentshould not apply to the prescribing of amedical device such as a contact lens. Wesought to determine the degree to whichcontact lens wearers have been informed ofthe risks associated with contact lens wear.

MethodsFollowing permission from the respectivemedical school deans, students at UK medicalschools were contacted through electronicmail by their respective university facultiesand asked if they wore or had worn contactlenses in the past year, to follow a link to aquestionnaire (fig 1) sited on the world wideweb.

ResultsIn all, 560 replies were received (table 1); 21%of students had received some verbal infor-mation regarding the risk of corneal ulcera-tion and 1% had received this in writtenform. Only 2% had been told of all of the fourstated risks; 12–14% of respondents wereaware of the risk of corneal ulcerationassociated with soft daily, continuous, andhard contact lens wear. Of those who knew ofthis risk 41%, 37%, and 58% underestimatedthe respective morbidity. Only 23% of thosewho think that they had been given sufficientadvice knew of the risk of corneal ulceration.Thirteen per cent of students reported thedevelopment of one or more complicationsassociated with their contact lens wear. Eightper cent of students stated that had they beenaware of the risks stated in the questionnairethey would not have chosen to wear contactlenses. 87% of students would still havechosen to wear a contact lens after readingthe associated risks.

CommentContact lenses are worn by millions of peopleworldwide. Although the risk of a complica-tion is small, it constitutes a significantpublic health problem when applied to alarge population. Individually it is necessaryto balance this risk against the benefitsoffered by contact lens wear. This can onlybe achieved with the provision of appropriateinformation.Although there are limitations with this

type of survey, it would appear that themajority of contact lens wearers, approxi-mately 80% (98% for all four of the stated

risks) of students surveyed, were unaware ofthe risks to which they were exposed.It is important that potential contact lens

wearers are fully informed to allow the bestchoice of refractive correction for them.Consumer comparison between the potentialadverse effects of various contact lens typesand wear patterns is important in thisdecision. This is emphasised here as 8% ofthose involved in the study would not havechosen to wear contact lenses had theyknown of the associated risks.Although there is no legal stipulation

regarding the provision of information whenprescribing a contact lens, providing informa-tion and obtaining informed consent formspart of good clinical practice. Althoughwritten consent is not evidence that informeddiscussion has taken place, written agree-ments can be used to provide a basis for theprocess.We would recommend that informed con-

sent be obtained when prescribing contactlenses and that this should be done with theaid of written information. Concern over theassociated risks should not deter people fromthe benefits of contact lens wear but they arean important consideration and consumershave a right to be informed.

A Roberts, A E Kaye, R A Kaye, K Tu, S B KayeSt Paul’s Eye Unit, 8Z Link, Royal Liverpool University

Hospital, Prescot Street, Liverpool L7 8XP, UK

Correspondence to: Stephen B Kaye, St Paul’s EyeUnit, 8Z Link, Royal Liverpool University Hospital,

Prescot Street, Liverpool L7 8XP, UK; [email protected]

doi: 10.1136/bjo.2004.062315

References

1 Schein OD, Glynn RJ, Poggio EC, et al. Therelative risk of ulcerative keratitis among users ofdaily wear and extended wear contact lenses.N Engl J Med 1989;321:773–8.

2 MacRae S, Herman C, Stulting RD. Corneal ulcerand adverse reaction rates in premarket contactlens studies. Am J Ophthalmol1991;111:457–65.

3 Poggio EC, Glynn RJ, Schien OD, et al. Theincidence of ulcerative keratitis among users ofdaily wear and extended wear soft contact lenses.N Engl J Med 1989;321:779–83.

4 Cheng KH, Leung SL, Hoekman HW, et al.Incidence of contact lens associated microbialkeratitis and its related morbidity. Lancet1999;354:181–5.

Q1 For how long have you been wearing contact lenses? 1–5 years 5–10 years >10 yearsQ2 Has your contact lens practitioner ever told you of the following risks which are associated with contact lens wear? Corneal ulcers with loss of vision Yes/No Blood vessel formation on the cornea Yes/No Corneal scarring with loss of vision Yes/No Drooping of the upper eye lid (ptosis) Yes/NoQ3 Has your contact lens practitioner ever given you written information on the following risks which are associated with contact lens wear? Corneal ulcers with loss of vision Yes/No Blood vessel formation on the cornea Yes/No Corneal scarring with loss of vision Yes/No Drooping of the upper eye lid (ptosis) Yes/NoQ4 Were you offered a choice of contact lens type by your contact lens practitioner? Yes/NoQ5 Do you receive your contact lenses from hospital (Yes/No) or high street (Yes/No) contact lens practitioner? Q6 Do you feel that the information given to you by your contact lens practitioner was sufficient? Yes/No/Unsure Q7 On a scale from one to five, with one being very well and five being not at all, how well do you think you understood the information you were given by your contact lens practitioner? 1–5Q8 Have you ever developed any of the following problems? Corneal ulcers Yes/No Blood vessel formation on the cornea Yes/No Corneal scarring Yes/No Drooping of the upper eye lid (ptosis) Yes/NoQ9 The estimated risk of corneal ulceration associated with contact lens wear is stated below. Please indicate if you were aware of these given risks. If the answer is yes please select whether you thought the risk was greater than stated, less than stated or the same as stated.

Q10 If you were not aware of these risks, would you still have worn contact lenses had you be given the above information? Yes/No

Soft daily wear lenses Soft continuous wear lenses Hard lenses

Risk

1/20001/5001/3000

Aware ofrisk?Yes/NoYes/NoYes/No

GreaterthanYes/NoYes/NoYes/No

Less than

Yes/NoYes/NoYes/No

Same as

Yes/NoYes/NoYes/No

Figure 1 Questionnaire.


782 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

Multiple use of single usesolutions: a dangerous practiceI am writing to raise an issue for routineophthalmic and optometry practice thatcontravenes the principles of clinical govern-ance and generates an avoidable infectionrisk to patients.My concern relates to single use preserva-

tive-free ophthalmic solutions, particularlythose used for diagnostic purposes in hospitaleye clinics and optometry practices. Theregulatory guidelines1 and the updated gui-dance note from the Royal PharmaceuticalSociety (www.rpsgb.org.uk/pdfs/ophprepguid.pdf) specify that products designed andlicensed for single use must be discardedimmediately after use.In practice, however, a recent study2 has

documented the routine misuse of single usesolutions (Minims), generating an unneces-sary risk of infection transferral betweenpatients. Single use solutions used multipletimes were shown to have a significantlyhigher rate of contamination than those thatwere used once then discarded (45% versus4%). An associated audit (Qureshi MA,personal communication, 2004) of hospitalsin the London region found that 86% of theophthalmologists surveyed admitted to mul-tiple uses of single use solutions betweendifferent patients.The study demonstrated the principle that

single use products could act as a vehicle fortransmission of micro-organisms present onthe practitioner’s hands or in the lid flora, ifshared between patients. We know thatgloves are typically not worn in outpatientophthalmic practice, and that hands are notalways washed between each and everypatient. We note that even though the eyedrop unit itself does not touch the patient’seye or surrounding area, the practitioneroften does touch the patient with his/herhands. Furthermore, the potential exists forinfection transfer without physical contact.Aside from the risk of infecting the patient,

the efficacy of the medication can be com-promised in the event of contamination.My own experience of ophthalmologists

and hospital pharmacists indicates that insome cases the interpretation of the term‘‘immediately’’ has been stretched beyondthe simple instruction to ‘‘use once anddiscard.’’ For example, some publishedpapers (Marchese et al3) refer to a, nowoutdated, EMEA instruction to use suchproducts within three hours of opening. TheEMEA has since replaced this with anupdated guideline4 indicating that the usagetime after opening is dependent on thenature of the product, yet this stillencourages the user to retain the productfor multiple use after first opening.Clinical governance requires that risks and

hazards to patients be reduced to as low alevel as possible. Single use products areavailable for this purpose and should be usedas intended and as licensed for one patientonly. Repeated use in the same patient over alimited period up to 24 hours after firstopening the product may be possible,depending on the product, and at the discre-tion of the clinician who must ensure that theproduct is not used on any other patient. Iurge your readers to take action as necessaryto ensure compliance.

Correspondence to: D V Seal, MD, Applied VisionResearch Centre, City University, Northampton

Square, London EC1V 0HB, UK;[email protected]

doi: 10.1136/bjo.2004.064535

References

1 Health Service Circular HSC (IS) 122. London:DHSS, 1975.

2 Wong R, Qureshi MA, Rowe C, et al. Dangers ofmultiple use of Minims preservative free eye dropsin applanation tonometry. 5th InternationalSymposium on Ocular Pharmacology andTherapeutics (ISOPT), 2004.

3 Marchese A, Bozzolasco M, Gualco L, et al.Evaluation of spontaneous contamination ofocular medications. Chemotherapy2001;47:304–8.

4 EMEA. Maximum shelf life for sterile products forhuman use after first opening or followingreconstitution. CPMP/QWP/159/96corr, 1998(www.emea.eu.int/pdfs/human/qwp/015996en.pdf).

Drug induced autoenucleationwith resultant chiasmal damageI read with interest the report of Tuwir et al1

of a case of autoenucleation with chiasmaldamage. They note the rarity of this dramaticself mutilating event, and state that therehave been only two other reports of contra-lateral field defect from chiasmal injury ininstances of autoenucleation.There have, however, been a number of

other reports of temporal field defect in theremaining eye following either self enuclea-tion or traumatic enucleation by an assailant.The consequences are the same regardless ofwho removes the eye. One of the two reportsTuwir et al cite, Parmar et al,2 was of a youngman whose eye was digitally enucleated byan assailant.In their 1984 review article on autoenu-

cleation, Krauss et al3 reported a 29 year oldwoman who had enucleated her right globecausing a complete temporal hemianopia inher remaining eye. They also reviewed themedical literature and found 19 cases ofbilateral self enucleation (or of self mutila-tion so severe as to require enucleation) and31 cases of unilateral self enucleation. Of the18 eyes where the length of the attached opticnerve was noted, six occurred at or near thechiasm. They also found seven cases oftraumatic avulsion of one eye resulting in atemporal hemianopia in the remaining eye.In 1996, Arkin et al4 reported a 25 year old

man with a temporal defect in his right eyeafter his left eye was pulled out by anassailant. A magnetic resonance image con-firmed damage to the chiasm. And in 2002,Dilly and Imes5 reported a 54 year oldschizophrenic man who removed his blindleft eye causing a complete temporal hemi-anopia in his remaining right eye. His MRIshowed avulsion of the lateral portion of hischiasm.It appears that even though acts of trau-

matic enucleation are rare, associated chias-mal injury is common.

Correspondence to: Richard K Imes, CPMC, USA;[email protected]

doi: 10.1136/bjo.2005.068502

References

1 Tuwir I, Chako E, Brosnahan D, et al. Druginduced autoenucleation with resultant chiasmaldamage. Br J Ophthalmol 2005;89:121.

2 Parmar B, Edmunds B, Plant G. Traumaticenucleation with chiasmal damage: magneticresonance image findings and response tosteroids. Br J Ophthalmol 2002;86:1317–18.

3 Krauss H, Yee R, Foos R. Autoenucleation. SurvOphthalmol 1984;29:179–87.

4 Arkin M, Rubin P, Bilyk J, et al. Anterior chiasmaloptic nerve avulsion. AJNR 1996;17:1777–81.

5 Dilly J, Imes R. Autoenucleation of a blind eye.J Neuro-ophthalmol 2001;21:30–1.

Table 1 Survey of 560 medical students who are contact lens wearers. All resultsare in percentages

Provision of information versus reported complications (%)Risk Verbal Written Complications*Ulcer 18 4 3BV 22 4 8Scar 21 4 4Ptosis 4 3 1Reported awareness of risk of corneal ulcerationSDL Aware UnawareGreater 18 17Same 41 18Less 41 65SCL Aware UnawareGreater 23 21Same 40 7Less 37 72HGP Aware UnawareGreater 40 29Same 2 0Less 58 71Knowledge of risk of corneal ulcer versus adequacy of advice

All Adequate advice Inadequate adviceSDL 14 23 21SCL 14 24 22RGP 12 23 17Percentage (%) who would wear contact lens with knowledge of risksYes 87No 8

*Actual complications reported by students. Blood vessel formation on the cornea (BV). Soft disposable(SDL), soft contact lens (SCL), rigid gas permeable (RGP), or hard contact lens (HGP).


MAILBOX

Accepted for publication 8 February 2005

PostScript 783


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

P-glycoprotein expression inretinoblastomaWe have read with interest the paper byKrishnakumar et al.1 We have studied 18children with retinoblastoma using immuno-histochemical detection of P-glycoprotein bythe mouse monoclonal antibody HYB-241 infrozen section tumour samples. LikeKrishnakumar et al, we found limited expres-sion of P-glycoprotein in retinoblastoma cellssince only four of the 18 samples werepositive. However, six of 18 samples had P-glycoprotein positivity in tumour associatedendothelial cells. Contrary to Krishnakumaret al, we had the opportunity of evaluatingchildren who received previous therapy withdrugs that can be affected by P-glycoproteinand no obvious difference in its expressionwas evident. Only one of the three tumoursamples coming from patients who had beenexposed to P-glycoprotein mediated che-motherapeutic agents (vincristine and/oretoposide) were positive in retinoblastomacells. In addition, of the six cases in which P-glycoprotein was detected in tumour asso-ciated endothelial cells, only one of themcame from a patient who had been exposedto P-glycoprotein mediated chemotherapeuticagents (etoposide). Also, one patient whosetumour and tumour associated endotheliumwere negative had endothelial cells in theoptic nerve that were positive. The patienthad not been exposed to P-glycoproteinmediated chemotherapy.There is little information regarding its

expression in normal eye blood vessels, but P-glycoprotein expression has been detected inthe normal human retinal pigment epithe-lium and post-laminar optic nerve,2 and it isabsent in the permeable microvessels of thechoroid and the ciliary process.3 P-glycopro-tein expression in endothelial cells of newlyformed capillaries induced by tumours hasalso been described.4 Its prognostic relevancein retinoblastoma is unknown, and it cannotbe determined from our data, since all ofthese eyes were already enucleated and nopatient had an extraocular relapse.The potential role of P-glycoprotein in

tumour associated endothelium creating ablood-tumour barrier is uncertain. Ourresults suggest that if cyclosporine is indeedimproving the efficacy of chemotherapy, itmay be acting by improving its deliverythrough P-glycoprotein expressing tumourassociated endothelial cells or through alter-ing pharmacokinetics of chemotherapeuticdrugs, rather than exclusively acting toreverse drug resistance at the level of thetumour cell.Our data do not support the hypothesis

that cyclosporine will be a useful adjunct tochemotherapy for the treatment of retino-blastoma via competitive inhibition of P-glycoprotein in tumour cells.

I J Dunkel, D H Abramson, C Cordon-CardoMemorial Sloan Kettering Cancer Center, Box 185,

1275 York Avenue New York, NY 10021, USA

G L Chantada, M T G de Davila, A C FandinoHospital JP Garrahan, Combate de los Pozos 1881,

C1245AAL, Buenos Aires, Argentina

Correspondence to: Dr Ira J Dunkel, Memorial SloanKettering Cancer Center, Box 185, 1275 York

Avenue, New York, NY 10021, USA;[email protected]

References

1 Krishnakumar S, Mallikarjuna K, Desai N, et al.Multidrug resistant proteins: P-glycoprotein andlung resistance protein expression inretinoblastoma. Br J Ophthalmol2004;88:1521–6.

2 Kennedy B, Mangini N. P Glycoproteinexpression in human retinal pigment epithelium.Mol Vis 2002;8:422–30.

3 Hofman P, Hoyng P, vander Werf F, et al. Lack ofblood-brain barrier properties in microvessels ofthe prelaminar optic nerve head. InvestOphthalmol Vis Sci 2001;42:895–901.

4 Camassei FD, Arancia G, Cianfriglia M, et al.Nephroblastoma: multidrug resistance P-glycoprotein expression in tumor cells andintratumoral capillary endothelial cells. Am J ClinPathol 2002;117:484–90.

Who is the author?We read with great interest Schwab’s answer1

to ‘‘Who is Ivan Schwab?’’2 We thank him forhis gracious response to our original letter,3

which suggested that readers would benefit ifthe BJO consistently published authors’ qua-lifications. Although Schwab and the editorsunderstood this to be a question of creden-tials to write about a subject, we maintainthat this issue has more to do with giving thereader perspective. An excellent example ofthe need to understand the authors’ perspec-tive comes from Schwab’s reply itself.Schwab made the presumption that we

were both ophthalmologists, which is, in fact,not correct. One of us (DP) is a medicalgraduate and currently a postgraduateresearch student. Given that the BJO doesnot publish authors’ qualifications, it is clearhow this occurred. We propose that it iscommon for readers to make inferencesabout the authors, as Schwab did. It isreasonable and logical to consider not onlythe information, but its source as well. Whenauthors’ qualifications are not explicitlystated, as in the BJO, readers can easily makeerroneous presumptions about the source.Does the BJO expect its readers to make nopresumptions about the backgrounds of theauthors? This would seem a difficult task forthe inquisitive mind. Does it matter thatreaders may make erroneous presumptionsabout the authors? We will leave this ques-tion to the reader.It is interesting to note that, on occasion,

the BJO does take care to give the reader morebackground on an author. For example, in anarticle on informed consent,4 the authors aregiven the designations of ‘‘solicitor’’ and‘‘ophthalmologist,’’ respectively. Similarly,the author of an article on professionalregulation,5 is given the designation‘‘Chairman of Committee on ProfessionalPerformance, General Medical Council,Portex Professor of Anaesthesia, Institute ofChild Health, London.’’ This supports ourview that it is important for the reader tounderstand the perspective of the author.Thus, the BJO’s approach does not appearconsistent.In answer to Schwab: indeed, we are

‘‘reading his essays, asking questions’’ andwill ‘‘stay tuned.’’1 In fact, we hope that indue course, the BJO will publish his series asa collection. We understand that the need toconserve space limits references. However,

rather than for verification, readers who havebeen stimulated by his writing would benefitfrom more references—perhaps online—tofurther their understanding. For example, inAugust 2003, Schwab referred to Land andNilsson’s book Animal Eyes (Oxford: OxfordUniversity Press, 2002). This is an excellentresource to improve the reader’s understand-ing and appreciation of his essays, particu-larly those describing invertebrate eyes.Finally, we note that other BJO readers

have not commented on the subject ofauthors’ qualifications. Perhaps, ironically,other readers think that they do not havethe credentials to comment.

D Papalkar, I C FrancisDepartment of Ophthalmology, Prince of Wales

Hospital, Randwick, NSW, Australia and theUniversity of NSW, Sydney, NSW, Australia

Correspondence to: Dr Ian C Francis, Suite 12,Chatswood Grove, 12–14 Malvern Avenue,

Chatswood 2067, Australia; [email protected]

doi: 10.1136/bjo.2005.069104

References

1 Schwab IR. In answer to ‘‘Who is Ivan Schwab? ’’Br J Ophthalmol, 2005;89:249.

2 Hoyt CS, Dick A, Bhisitkul B. Who is IvanSchwab? Br J Ophthalmol 2004;88:1106.

3 Papalkar D, Francis IC. Authors’ qualificationsand the BJO. Br J Ophthalmol 2004;88:1227.

4 Teuten B, Taylor D. ‘‘Don’t worry my good man—you won’t understand our medical talk’’: consentto treatment today. Br J Ophthalmol2001;85:894–6.

5 Hatch D. Professionally led regulation inmedicine. Br J Ophthalmol 2001;85:513–15.

Vitrectomy with and withoutscleral buckle for inferior retinaldetachmentIn the article presented by Wickham andassociates,1 the authors compared vitrectomyand gas for treating inferior break retinaldetachments with vitrectomy, gas and scleralbuckle. The study showed no significantdifference in the final outcome between thetwo groups. While vitrectomy and gas forinferior break retinal detachments appearspromising, there are several issues that wewould like to raise.Firstly, the surgery was performed by a

registrar, fellow, or consultant. These sur-geons may have varying degrees of experi-ence and the inconsistency may affect therate of successful surgical outcome. Secondly,additional tears were treated with cryother-apy or laser. As shown by Bonnet et al,2 thepostoperative proliferative vitreoretinopathy(PVR) rate could be as high as 25.8% inpatients treated with cryotherapy comparedto 2.2% in the laser group. It is unclear whatthe relative distribution of patients whounderwent cryotherapy in the two groupswas and this may have been a confoundingfactor in the study. Thirdly, patients under-went an air/gas exchange with either SF6 orC3F8. As C3F8 had a much longer duration oftamponade than SF6, the use of one agentover another may have led to a difference inthe success rate.The study excluded patients with PVR

grade C. However, for those with grade A orB, a scleral buckle was planned before theoperation. This could lead to a selection biaswhere potentially more difficult cases were

Disclosure: There is no financial interest of the authorswith any of the drugs and devices mentioned.

doi: 10.1136/bjo.2005.068536

Accepted 8 February 2005


784 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

scheduled into the scleral buckle group. Thismay be a contributing factor for a higher rateof postoperative PVR (20%) and epiretinalmembrane formation in this group, comparedto a rate of 5–10% reported previously.3 4 Theunderlying vitreoretinal pathology ratherthan the placement of the scleral bucklemay have been a major reason behind thehigh PVR rate noted in this group.The authors stated that the main reasons

for performing vitrectomy and gas withoutscleral buckle was to avoid the possiblecomplications of scleral buckle—namely,longer operating time,5 exposure, refractivechange, diplopia, and anterior segmentischaemia.6–10 Perhaps, in the interest ofreaders, the authors can provide us with theinformation if any of these complicationsdeveloped during the study.The high rate of final reattachment

reported in the study is encouraging. Webelieve that vitrectomy and gas alone is aneffective method to treat selected cases ofretinal detachments with inferior retinalbreaks. A controlled, randomised, prospectivestudy, comparing the outcome in properlymatched groups and with meticulous atten-tion to surgical methods11 will help addresssome of the above issues and help elucidatefurther if the procedure without the use ofscleral buckle will benefit patients withinferior break retinal detachment.We would like to commend the authors for

conducting this very nice study on animportant topic that may provide a betteralternative treatment. We wish that theissues that we raise will help broaden thediscussion on the topic.

Y Y Y Kwong, C W Tsang, W W Lai, D S C LamHong Kong, People’s Republic of China

Correspondence to: Professor Dennis Lam, 3/F, HongKong Eye Hospital, 147K Argyle Street, Kowloon,

Hong Kong; [email protected]

doi: 10.1136/bjo.2005.068528

References

1 Wickham L, Connor M, Aylward GW. Vitrectomyand gas for inferior break retinal detachments:are the results comparable to vitrectomy, gas, andscleral buckle? Br J Ophthalmol2004;88:1376–9.

2 Bonnet M, Guenoun S. Surgical risk factors forsevere postoperative proliferativevitreoretinopathy in retinal detachment with gradeB PVR. Graefes Arch Clin Exp Ophthalmol1995;233:789–91.

3 Charteris DG, Sethi CS, Lewis GP, et al.Proliferative vitreoretinopathy. V Developments inadjunctive treatment and retinal pathology. Eye2002;16:369–74.

4 The Retina Society Terminology Committee. Theclassification of retinal detachment withproliferative vitreoretinopathy. Ophthalmology1983;90:121–5.

5 Hakin KN, Lavin MJ, Leaver PK. Primaryvitrectomy for rhegmatogenous retinaldetachment. Graefes Arch Clin Exp Ophthalmol1993;231:344–6.

6 Findall RJ, Norton EW, Curtin, et al. Reduction ofextrusion and infection following episcleralsilicone implants and cryopexy in retinaldetachment surgery. Am J Ophthalmol1971;71:835–7.

7 Hayashi H, Hayashi K, Nakao F, et al. Cornealshape changes after scleral buckling surgery.Ophthalmology 1997;104:831–7.

8 Domniz Y, Cahana M, Avni I. Corneal surfacechanges after pars plana vitrectomy and scleralbuckling surgery. J Cataract Refract Surg2001;27:868–72.

9 Fison PN, Chignell AH. Diplopia after retinaldetachment surgery. Br J Ophthalmol1987;71:521–5.

10 Kwartz J, Charles S, Mc Cormack P, et al.Anterior segment ischaemia following segmentalscleral buckling. Br J Ophthalmol1994;78:409–10.

11 Leaver P. Expanding the role of vitrectomy inretinal reattachment surgery. Br J Ophthalmol1993;77:197.

Charles Bonnet syndrome inpatients with glaucoma and goodacuityWe read with interest the paper by Tan et al1

on Charles Bonnet syndrome (CBS) in Asianpatients. Their finding of a lower CBSprevalence than European or NorthAmerican surveys demands further investiga-tion, although this may reflect the stringentcriteria of hallucination complexity they usedin making the diagnosis (thus excluding thecommonest CBS hallucinations of colouredblobs and grid-like ‘‘tesselloptic’’ patterns2 3)and, as pointed out in the accompanyingeditorial comment, the relatively low preva-lence of macular disease in their cohort.However, it is not this aspect of the reportwe found most intriguing—it was the obser-vation that CBS occurred with good acuity. Infact, three of the four CBS patients describedhad a degree of impairment which placedthem at risk for CBS (best eye acuity 0.3 orworse4). It is the remaining patient (patientthree, a 72 year old man) who is of particularimportance as his relative preservation ofacuity bilaterally (20/30 RE, 20/40 LE) chal-lenges the view that significant acuity loss isa prerequisite for ‘‘ophthalmological’’ visualhallucinations. This case mirrors fourpatients we have recently studied with CBSsecondary to glaucoma and bilaterally goodacuity. We describe the cases below and offera pathophysiological mechanism for theassociation.In one sense, the finding that CBS occurs

with preserved acuity is hardly novel. As citedby Tan et al,1 several previous reports havefound such an association. However, all is notas it seems, the term CBS being used indifferent ways by different authors. Some usethe term to describe visual hallucinationswith insight, irrespective of the presence ofeye disease, age or clinical context.5 6 Othersuse the term to describe the association ofvisual hallucinations with age and intactcognition, without reference to eye diseaseor hallucination phenomenology.7 8 Underthese definitions it is hardly surprising thata patient with ‘‘CBS’’ has preserved acuity,the patients in these studies having a diverserange of conditions from delirium toParkinson’s disease and beyond. In contrast,ophthalmologists and neurologists have usedCBS to emphasise eye or visual pathwaydisease, with the phenomenology of thehallucinations and age being of secondaryimportance.9 10 Although each definition ofCBS has its merits, the ophthalmologicaldefinition reminds us best of Bonnet’soriginal description and helps characterise adistinctive subgroup of visually hallucinatingpatients with predicable prognosis and spe-cific pathophysiology.2 11 However, even CBSas defined ophthalmologically carries with itan inherent ambiguity: is it eye disease itselfor the loss of acuity that is the importantfactor? The consistent finding of acuity lossas a risk factor4 12 13 suggests the latter, or atleast that the central retina has a key role in

the underlying pathophysiological mechan-ism.As part of a larger study into the visual

phenomenology of CBS, we have recruitedfour patients with advanced glaucoma (threePOAG and one chronic narrow angle) butpreserved visual acuity. The age range of thepatients was 81–91 years, three men and onewoman. Their visual acuities ranged from 6/6to 6/12 monocularly with all patients having6/9 or better in their better eye. All hadextensive field defects bilaterally and cup todisc ratios of 0.8 or greater in both eyes. Twopatients had bilateral trabeculectomies nowoff treatment, one was on bimatoprost andTrusopt to both eyes and one on timolol0.25% to both eyes. Two patients werebilaterally pseudophakic. The patient withchronic narrow angle glaucoma had previoussurgical iridectomies. The duration of theirhallucinations ranged from 6 months to6 years. Three patients hallucinated in colourand one in black and white. The mostcommon hallucination was of tessellopsia2

experienced by all the patients, with twopatients seeing, in addition, formed buildingsand two patients, letter-like shapes. Therewere also single reports of hallucinations ofgroups of people, animals, branching shapes(dendropsia2) and one patient describedvisual allesthesia.14 In three of the patientsthe hallucinations encompassed the entirevisual field, in the fourth they were restrictedto the visual field defect. None had halluci-nations in other sensory modalities and allhad insight into the nature of the experi-ences. The phenomenology of the hallucina-tions and the relative frequency of thedifferent hallucination categories are consis-tent with previous descriptions of ophthal-mologically defined CBS.2 3 Non-ophthalmological causes of visual hallucina-tions2 3 were excluded. As far as we are awarethis is the largest case series of patients withvisual hallucinations secondary to eye diseaseand bilaterally preserved visual acuity yet tobe reported.Current aetiological theories of CBS

emphasise the importance of deafferenta-tion15 (both ‘‘physiological’’ through ganglioncell loss and ‘‘functional,’’ for example,related to blindfolding or cataract), the lossof visual input resulting in a change incortical excitability.2 Although it has beenassumed that deafferentation of sufficientseverity to precipitate CBS implies a conse-quent loss of acuity, our cases and that of Tanet al1 suggest otherwise. Patients withadvanced glaucoma can have a significantdegree of ganglion cell loss and consequentphysiological deafferentation without a lossof acuity, placing them at risk for CBS. Thiscontrasts with age related macular diseasewhere the loss of central retinal ganglion cellsleads, indirectly, to an association of CBSwith acuity loss. We conclude that reducedacuity is not a necessary prerequisite forophthalmologically defined CBS and thatophthalmologists should be aware thatpatients with preserved acuity but significantdeafferenting ocular disease are at risk of thesyndrome.

S A Madill, D H FfytcheEye Department, King’s College Hospital, Denmark

Hill, London SE5 9RS, UK

Correspondence to: Mr Stephen Madill, EyeDepartment, King’s College Hospital, Denmark Hill,

London SE5 9RS, UK; [email protected]

doi: 10.1136/bjo.2005.066027


PostScript 785


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

References

1 Tan CSH, Lim VSY, Ho DYM, et al. Charles Bonnetsyndrome in Asian patients in a tertiaryophthalmic centre. Br J Ophthalmol2004;88:1325–9.

2 ffytche DH, Howard RJ. The perceptualconsequences of visual loss: positive pathologiesof vision. Brain 1999;122:1247–60.

3 Santhouse AM, Howard RJ, ffytche DH. Visualhallucinatory syndromes and the anatomy of thevisual brain. Brain 2000;123:2055–64.

4 Teunisse RJ, Cruysberg JR, Verbeek AL, et al. TheCharles Bonnet syndrome: a large prospectivestudy in the Netherlands. Br J Psychiatry1995;166:254–7.

5 Damas-Mora J, Skelton-Robinson M, Jenner FA.The Charles Bonnet syndrome in perspective.Psychol Med 1982;12:251–61.

6 Gold K, Rabins PV. Isolated visual hallucinationsand the Charles Bonnet syndrome: a review of theliterature and presentation of six cases. ComprPsychiatry 1989;30:90–8.

7 Podoll K, Osterheider M, Noth J. Das CharlesBonnet-Syndrom. Fortschr. Neurol Psychiat1989;57:43–60.

8 De Morsier G. Le syndrome de Charles Bonnet:hallucinations visuelles des vieillards sansdeficience mentale. Annales Medico-Psychologiques 1967;125:677–702.

9 Manford M, Andermann F. Complex visualhallucinations. Clinical and neurobiologicalinsights. Brain 1998;121:1819–40.

10 Menon GJ, Rahman I, Menon SJ, et al. Complexvisual hallucinations in the visually impaired: theCharles Bonnet syndrome. Surv Ophthalmol2003;48:58–72.

11 ffytche DH. Visual hallucination and illusiondisorders: a clinical guide. Adv Clin NeurosciRehab 2004;4:16–18.

12 Holroyd S, Rabins PV, Finkelstein D, et al. Visualhallucinations in patients with maculardegeneration. Am J Psychiatry1992;149:1701–6.

13 Scott IU, Schein OD, Feuer WJ, et al. Visualhallucinations in patients with retinal disease.Am J Ophthalmol 2001;131:590–8.

14 Girkin CA, Miller NR. Central disorders of visionin humans. Surv Ophthalmol 2001;45:379–405.

15 Burke W. The neural basis of Charles Bonnethallucinations: a hypothesis. J Neurol NeurosurgPsychiatry 2002;73:535–41.

Author’s replyI thank Madill and Ffytche for their interest-ing comments on our paper1 and am gratefulfor the opportunity to respond to some of theissues raised. The first relates to possiblevariations in the prevalence rate of CharlesBonnet syndrome (CBS) in different popula-tions, on which there is currently very littlesubstantive data. We agree that additionalpopulation based studies are necessary todetermine whether the differences foundbetween ethnic groups are consistent. Wewish to highlight that our findings1 aresupported by another recently publishedstudy of 1000 Japanese patients by Shiraishiet al2 who reported a prevalence rate of 0.5%compared to 0.4% in our study population.Since the methodology and diagnostic criteriaused in that study are very similar to ours, thecomparable prevalence rates reported in bothstudies may be a true reflection of a lowerprevalence of CBS in Asians compared toWestern populations.We would like to point out that the large

variation in prevalence rates of CBS in thevarious cross sectional studies may beexplained by the characteristics of the differ-ent populations being screened, especiallythe degree of visual impairment. Whenpatients were drawn from either a general

ophthalmology1–3 or general medical4 clinic,presumably comprising patients with rela-tively good visual acuity, the prevalence ratesranged from 0.4–2%. In patients with poorervisual acuity3 5–8 the prevalence rates variedfrom 11–12.3%. This observation suggeststhat poor visual acuity may be a factor thatincreases the risk of developing CBS. Indeed,Teunisse et al showed that the risk of CBS washigher in patients with a logMAR visualacuity of worse than 0.33 and Holroyd et alalso found an association with bilateral visualacuity of 20/60 or less.9

Nevertheless, even if reduced visual acuityis indeed a risk factor for the development ofCBS, we agree that a reduction in visualacuity is not essential and that CBS can occurin patients with relatively good visual acuity.Gold et al thought that reduced visual acuityis not a prerequisite for the diagnosis of CBS10

and Holroyd et al suggested that it may be thepresence of visual impairment and not theseverity of visual loss that predisposes toCBS.4 Schultz and Melzack, in an excellentreview of the cases reported in the literature,found that the visual acuities of patientsdiagnosed with CBS ranged from normal tono light perception.11 It is becoming apparentthat CBS can develop in patients with visualfield defects but otherwise good centralacuity. The patients described by Madill andFfytche and our own patient (patient 3) hadglaucomatous visual field defects. A recentpaper by Freiman et al12 documented thepresence of hallucinations manifestingwithin visual field defects that occurredfollowing neurosurgery. Two interestingpoints can be noted from Freiman et al’sseries: firstly, the hallucinations were con-fined to the area of visual loss; secondly, theybegan soon after the surgery and resolvedwithin days to 6 months after their onset.4 Incontrast, in three of four patients reported byMadill and Ffytche as well as our ownpatient, the hallucinations were not restrictedto the visual field defect. The relation of theonset of CBS symptoms to the developmentof visual field defects as well as the localisa-tion of hallucinations within visual fields areissues that warrant further study as they mayprovide additional insight into the pathophy-siology of CBS.

Correspondence to: Colin S H Tan, The Eye Institute,National Healthcare Group, Singapore;

[email protected]

doi: 10.1136/bjo.2005.068510

References

1 Tan CSH, Lim VSY, Ho DYM, et al. Charles Bonnetsyndrome in Asian patients in a tertiaryophthalmic centre. Br J Ophthalmol2004;88:1325–9.

2 Shiraishi Y, Terao T, Ibi K, et al. The rarity ofCharles Bonnet syndrome. J Psychiatr Res2004;38:207–13.

3 Teunisse RJ, Cruysberg JR, Verbeek AL, et al. TheCharles Bonnet syndrome: a large prospectivestudy in the Netherlands. Br J Psychiatry1995;166:254–7.

4 Holroyd S, Rabins PV, Finkelstein D, et al. Visualhallucinations in patients from an ophthalmologyclinic and medical clinic population. J Nerv MentDis 1994;182:273–6.

5 Brown GC, Murphy RP. Visual symptomsassociated with choroidal neovascularization.Photopsias and the Charles Bonnet syndrome.Arch Ophthalmol 1992;110:1251–6.

6 Olbrich HM, Engelmeier MP, Pauleikhoff D, et al.Visual hallucinations in ophthalmology. GraefesArch Clin Exp Ophthalmol 1987;225:217–20.

7 Teunisse RJ, Cruysberg JR, Hoefnagels WH, et al.Visual hallucinations in psychologically normalpeople: Charles Bonnet’s syndrome. Lancet1996;347:794–7.

8 Nesher R, Nesher G, Epstein E, et al. CharlesBonnet syndrome in glaucoma patients with lowvision. J Glaucoma 2001;10:396–400.

9 Holroyd S, Rabins PV, Finkelstein D, et al. Visualhallucinations in patients with maculardegeneration. Am J Psychiatry1992;149:1701–6.

10 Gold K, Rabins PV. Isolated visual hallucinationsand the Charles Bonnet syndrome: a review of theliterature and presentation of six cases. ComprPsychiatry 1989;30:90–8.

11 Schultz G, Melzack R. The Charles Bonnetsyndrome: ‘phantom visual images’. Perception1991;20:809–25.

12 Freiman TM, Surges R, Vougioukas VI, et al.Complex visual hallucinations (Charles Bonnetsyndrome) in visual field defects followingcerebral surgery. Report of four cases.J Neurosurg 2004;101:846–53.

Tight necktie, intraocularpressure, and intracranialpressureI would like to congratulate Theelen et al fortheir recent article on impact factors onintraocular pressure measurements inhealthy subjects,1 and I would like to add athought. As Theelen and colleagues point outwith reference to the literature,2 3 increasedpressure in the jugular vein leads to increasedbrain pressure, and by an increase in theepiscleral venous pressure, to an elevation ofintraocular pressure. Correspondingly, in aprevious study by Teng and associates, it wasdiscussed that a tight necktie may increaseintraocular pressure by an increased jugularvein pressure and could affect the diagnosisand management of glaucoma.4 It may betaken into account, however, that the brainpressure and pressure in the cerebrospinalfluid space surrounding the retrobulbar partof the optic nerve are the counter-pressureagainst the intraocular pressure across thelamina cribrosa.5 If the cerebrospinal fluidspace pressure is elevated (as a result ofincreased jugular vein pressure), the intra-ocular pressure may also be allowed to beelevated so that the trans-lamina cribrosapressure difference may remain constant.Independently of the question of whether atight necktie may or may not increaseintraocular pressure, one may assume thatif the intraocular pressure gets higherbecause of an increased jugular vein pressure,it may, at least partially, be balanced by anincrease in brain pressure, without increasingthe risk for glaucoma.

Correspondence to: J B Jonas, Department ofOphthalmology, Theodor-Kutzer-Ufer 1-3 Mannheim,

68167, Germany;[email protected]

doi: 10.1136/bjo.2005.068544

References

1 Theelen T, Meulendijks CFM, Geurts DEM, et al.Impact factors on intraocular pressuremeasurements in healthy subjects. Br J Ophthalmol2004;88:1510–1.

2 Mavrocordatos P, Bissonnette B, Ravussin P.Effects of neck position and head elevation onintracranial pressure in anaesthetized




786 PostScript


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

neurosurgical patients—preliminary results.J Neurosurg Anesthesiol 2000;12:10–4.

3 Toole JF. Effects of change of head limb and bodyposition on cephalic circulation. N Engl J Med1968;279:307–11.

4 Teng C, Gurses-Ozden R, Liebmann JM, et al.Effect of a tight necktie on intraocular pressure.Br J Ophthalmology 2003;87:946–94.

5 Jonas JB, Berenshtein E, Holbach L. Laminacribrosa thickness and spatial relationshipsbetween intraocular space and cerebrospinal fluidspace in highly myopic eyes. Invest OphthalmolVis Sci 2004;45:2660–5.

Authors’ replyWe thank Dr Jonas for his interest in ourwork and his important supplements aboutpathophysiological aspects regarding neckcirculation and intraocular pressure. As theoptic nerve head is in close contact with bothintraocular and subarachnoidal space, onemay imagine a disc protecting balance inchronically elevated jugular pressure.1 Shortterm fluctuations of cervical circulation, aspresent in our recent study, may not facilitatethis guarding benefit.2 Thus, repeated hugevariations of jugular pressure can presumablybe a risk factor for glaucomatous optic nervehead damage. Obviously, this is not the casein slit lamp adapted measurement of intra-ocular pressure. Nevertheless, the patient’sneck position during tonometry should betaken into consideration to improve theinterpretation of intraocular pressure mea-surements.

T Theelen, C F M MeulendijksRadboud University Nijmegen Medical Centre, Philipsvan Leijdenlaan 15 Nijmegen, 6525 EX, Netherlands

Correspondence to: T Theelen, Radboud UniversityNijmegen Medical Centre, Philips van Leijdenlaan 15

Nijmegen, 6525 EX, Netherlands;[email protected]

doi: 10.1136/bjo.2005.069757

References

1 Jonas JB, Berenshtein E, Holbach L. Laminacribrosa thickness and spatial relationshipsbetween intraocular space and cerebrospinal fluidspace in highly myopic eyes. Invest OphthalmolVis Sci 2004;45:2660–5.

2 Theelen T, Meulendijks CFM, Geurts DEM, et al.Impact factors on intraocular pressuremeasurements in healthy subjects. Br J Ophthalmol2004;88:1510–11.


CORRECTION

doi: 10.1136/bjo.2005.48777corr1

In the paper titled Prevalence and causes ofblindness and low vision in leprosy villagesof north eastern Nigeria (Br J Ophthalmol2005;89:417-9) the author has notified us ofan error. The second-last sentence in thesecond column on page 417 should read,‘Definitions of blindness and visual impair-ment were: VA > 6/18, normal; 6/60 ( VA ,6/18, visual impairment; 3/60 ( VA , 6/60,severe visual impairment…’.

NOTICES

Keppel Street, London WC1E 7HT, UK (tel:+44 (0)20 7612 7964; email: [email protected]; online edition: www.jceh.co.uk). Annual subscription (4 issues) UK £28/US$45. Free to developing country applicants.

EVER 2005 meetingThis will take place on 5–8 October 2005 inVilamoura, Portugal. For further detailsplease contact: Christy Lacroix, EVERSecretary, Kapucijnenover 33, B-3000Leuven, Belgium (tel: +32 (0)16 233 849;fax +32 (0)16 234 097; email:[email protected]).

World Ophthalmology Congress2006 – BrazilThe World Ophthalmology Congress (whichis replacing the International Congress ofOphthalmology) is meeting in February 2006in San Paulo, Brazil.For further information on the congress

and committees, scientific program and coor-dinators of different areas are available at thecongress website www.ophthalmology2006.com.br

Worldwide clinical trials for newtechnique for early detection of eyediseaseA unique new non-invasive technique forhigh resolution optical imaging of the eye isreceiving global acclaim. By combining twohigh-resolution imaging technologies, thenew technique provides doctors with 3-Dimages of the retina, macula and the opticnerve.For more information, contact the Media

Office on 01227 823581/823100 or [email protected] News releases canalso be found at: http://www.kent.ac.uk/news

Trachoma controlThe latest issue of Community Eye Health(No 52) discusses new developments in thecontrol of trachoma. For further informationplease contact: Journal of Community EyeHealth, International Resource Centre, Inter-national Centre for Eye Health, Departmentof Infectious and Tropical Diseases, LondonSchool of Hygiene and Tropical Medicine,

PostScript 787


copyright. on N



j.com/

Br J O


ay 2005. Dow

nloaded from

http://bjo.bmj.com/

Documents

PostScript - British Journal of Ophthalmology · adnexa. Int Ophthalmol Clin 1997;37:39–59. 3 Knowles DM, Jakobiek FA, McNally L, et al. Lymphoid lyperplasia and malignant lymphoma