1

GAFFIFactSheet

FungalKeratitis

Keratitis is an infectionof thenormally transparent corneaof theeye,which causes

ulcerationandgradualopacification.Itmaybecausedbybacteriaorfungi(orchemical

injury)andisthemaincauseofunilateralcornealscarring1,2.Over100differentfungi

havecausedfungalkeratitisandnewpathogensareregularlydescribed3,howeverthe

common causative agents are Fusarium spp., Aspergillus flavus and fumigatus and

Candida albicans3,4. The condition is most prevalent in tropical and subtropical

locations. Fungi have been account for 20–60% of all culture-positive corneal

infectionsintheseclimates(Fig1).Thishighlydamagingcornealinfectionoftenleads

topermanentblindnessandsometimeseyeloss5.

Figure1.Proportionofmicrobialkeratitiscausedbyfungi5.Cornealabrasionorsignificanttraumafromanytypeofplantororganicmaterialare

the most common predisposing factors6. Other risk factors include

immunocompromise (including exposure to local or systemic corticosteroids),

diabetes,HIVinfection7,impairedtearing,incompleteeyelidclosureandpoorhygiene

practice in thosewhouse contact lenses. Seasonal variations in incidence have also

beendescribed1.

GLOBAL ACTION

FUND FOR

FUNGAL INFECTIONS

GLOBAL ACTION

FUND FOR

FUNGAL INFECTIONS

OLD VERSION

DARKER AREAS AND TEXT FIT WITHIN CIRCLE

SMALLER VERSION (ALSO TO BE USED AS MAIN

LOGO IN THE FUTURE)

2

Incidence

The annual global incidence of fungal keratitis is estimated at 1,051,800 (736,250-

1,367,300) cases5. The highest estimated incidences are in Asia and Africa, and the

lowest inEurope(Fig2)5.Thiscouldbeanunderestimatebecausesomesamplesare

microscopynegativeandnotculturedforfungi5.

Figure2.Publishedannualincidenceoffungalkeratitisindifferentcountries5.

A statistically significant correlationhasbeen foundbetweenGrossNational Income

(GNI) andaetiologyofmicrobial keratitis. Fungal keratitis is associatedwith lowGNI

countries8. In 2002, a government report from India estimated that keratitis

accounted for 9% of cases of blindness in India9. In Ugandan children with visual

impairment, visual loss after corneal ulceration was responsible for nearly 25% of

cases10.

Youngadultsarepredominantlyatrisk,withmenmoreoftenaffectedthanwomen.In

one series nearly 4% of cases were found in children11. The annual incidence of

microbial keratitis in contact lens wearers is about 10-fold higher than non-contact

lenswearers at 1.2-1,304/10,000, depending on the typeof lens, overnight use and

thequalityof lens care8,12. The rateofHIV infection in thosepresentingwith fungal

keratitisinTanzaniawastwicethedocumentedrateintheadultpopulation7,asseen

byotherworkers13.

3

Clinicalpresentation

The eye exhibits signs of inflammation:

injection, an anterior chamber reaction

and, possibly a hypopyon (pus in the

anterior chamber)2,3. Both bacterial and

fungal ulcers may present with a large

area of central necrosis (Fig 3) and a

significant hypopyon. Serrated (as

opposed to well demarcated) margins,

raisedsloughandcolourationotherthanyellowarestatisticallymoreoftenassociated

withafungalcause,whereas immunering,keraticprecipitates,perineural infiltrates,

endothelialplaque,andflareorcells intheanteriorchamberarenot14.Aprobability

score of 89% likelihood that the infection is fungal if serrated, feathery infiltrate

margins,andraisedslough(surfaceprofile)arepresent,andfibrinisabsentfromthe

anteriorchamber5,14.

Diagnosis

Diagnosisoffungalkeratitis isslowandcomplicated.Confirmationofthediagnosis is

made from corneal scrapings or biopsy, bymicroscopy and culture3. The procedure

requiresthattheeyebeanaesthetisedwithlocalanaestheticeyedrops.Ametalblade

isthenusedtocollectmaterialasepticallyfromthebaseandmarginoftheulcerunder

directvisionthroughaslit-lamp.Asfungigenerallypenetratedeepintothecornea,the

yieldof fungiobtainedusingswabs isusually inadequatetoconfirmadiagnosis.The

materialisthentransferredtoacleanglassmicroscopeslide,floodedwithpotassium

hydroxideandexaminedforfungalelementsbylightmicroscopy.Directmicroscopyof

cornealsmearsallowsthecliniciantorapidlydifferentiatebetweenafungalinfection

andothertypesofmicrobialkeratitisandisconsideredthegoldstandardfordiagnosis

of fungal infection.Thesensitivity fordetectingfungalkeratitishasbeenreportedto

be61–94%usingpotassiumhydroxide,85%using lactophenolblue,but just36–50%

usingatraditionalGramstain5.Calcofluor-whiteissaidtobeamainstayofdiagnosis,

Figure 3. An example of severeAspergilluskeratitis

4

andwhen combinedwithpotassiumhydroxide stains, sensitivityhasbeen shown to

riseto98·3%5.

Itisnotpossibletodifferentiatebetweengeneraandspeciesoffungionthebasisof

microscopicexaminationofthecornealsmearpreparationalone.Forthisreason,itis

advisedthatbothmicroscopyandculturearedonewheneverpossible.

Samples should be cultured on bacterial and fungalmedia3,5. Blood agar, chocolate

agar,andSabourauddextroseagarare inoculatedwithcornealscrapematerialusing

C-shaped streaks, because of the very small size of the inoculum, and only colony

growth within these parameters are regarded significant. Fungal growth is typically

slow, taking 48 hours to 10 days to become visible. Due to the diversity of fungi

cultured from cases of fungal keratitis, examination of cultures by a specialist

mycologist is typically necessary to identify the cultures3. Fusarium species are the

most common (Fig 4) , followedbyAspergillus spp. andCandida spp. Togetherwith

Penicilliumspp.,Alternariaspp.,Paecilomycesspp.,Curvulariaspp.andBipolarisspp.,

these three pathogenic species account for about 90% of cases, with rare fungi

(sometimes unidentified) comprising the remainder.Many cultures are negative for

bacteriaandfungi,sometimesbecauseofpriorantimicrobialtherapy.Culturenegative

microbialkeratitisisestimatedataround40%5.Thismeansahugeunderestimationof

fungalkeratitisburdenandanurgentneedtoimprovethediagnosticapproach,make

itsimplerandimprovesensitivity.

Fusariumcultureandmicroscopyofthecultureshowingsickle-shapedmacroconidia

5

Otherapproachesthatcanbehelpfulfor

diagnosis are molecular methods and

confocal microscopy but are technically

andcostdemanding15.Themoleculartool

of choice is PCR, which only requires a

small quantity of sample. PCR has been

shown to have high sensitivity and

specificity when compared with smear

stains and culture. PCR positively

identifies the causative fungal species in

92·6%of cases5. However, the technique

iscurrentlyoflimiteduseinlowresource

settings,wheretheburdenofdiseaseisgreatest.

Point of care testing of this disease would dramatically improve patient outcomes.

Slow diagnosis, including referral from one facility to another, worsens visual

outcome16.

Treatment

Responses to topical antifungal therapy are reasonable, with 75% of corneas not

severely affected and 60% of those severely affected being cured by topical 5%

natamycin (Primaricin)17,18.Other therapiesproduce similar response rates, although

natamycin is superior to voriconazole18. Advanced disease on presentation is

associatedwithworseoutcomes.

Natamycin 5% eye drops are not registered as treatment inmost countries, and all

suppliersareinIndia.Natamycineyedropsarenowlisted(2017)asaWHOEssential

Medicine.Thefollowingalternativeantifungaleyedropshavebeenusedwithvariable

successrates:amphotericin0.15-0.3%,flucytosine1%,econazole1%,miconazole1%,

clotrimazole1%,itraconazole1%,fluconazole1%andvoriconazole1-2%,caspofungin

0.5%3. Oral itraconazole and voriconazolemay be useful in some patients. It is not

Figure 5. Venn diagram showing howthe diagnosis of fungal keratitis wasmadeinLondon15

6

clear whether intrastromal or

subconjunctival antifungal injections

contributetosuccessoftreatmentof

fungalkeratitis.

Globalavailabilityofnatamycin5%at

affordablecostisneeded.

Surgery is sometimes required in

patients who fail to respond to

medical therapy or where there is a

threatofocularperforationortheformationofadescemetocoele.Surgeryshouldbe

preceded by medical therapy for as long as possible. Surgical procedures include

debridementorlamellarkeratectomy,formationofaconjunctivalflapoveraseverely

ulcerated area of the cornea (in an attempt to save the eyeball), or penetrating

keratoplasty if a donor cornea is available. In patients withmalignant glaucoma, to

restoredrainageofaqueoushumour,iridectomy,lensexcisionoranteriorvitrectomy

may be necessary. In intractable cases, with perforation of the eye, removal of the

eyeball(evisceration)isrequired.

Outcome

Ithasbeenestimatedthat95,000–115,000eyesaresurgicallyremovedeachyearasa

directresultoffungalkeratitis5.Incountrieswhereeyecareissuboptimal,thelossof

eyeswill probably be greater.Using outcomedata from the Pakistan study for low-

income andmiddle-income countries, it is predicted that 610,800 eyeswill go blind

becauseoffungalkeratitiseachyear5.

Opportunities to reduce Global Disease Burden and improve patient

outcomes:

Anumberoffeasibleinitiatives, ifwidelyimplemented,couldhaveaverysubstantial

impactonreducingtheglobalfungalkeratitisdiseaseburden:

1. Encouragetheperformanceofepidemiologicalstudies includingestimationof

DALYs,tounderstandtherealburdenoffungalkeratitis

Figure 6. Outcomes from fungal keratitis withnatamycin5%eyedrops.

7

2. Provide training and availability in classical diagnostic procedures including

sampling,culturetechniquesandfungalspeciesidentification

3. Optimize use of antifungal therapy in resource limited settings through

promotingaglobalapproachtotheprevention,diagnosisandmanagementof

microbialkeratitis

4. Develop a point of care antigen test able to differentiates bacterial infection

fromfungalinfection,

5. Investigatethevalueofintroducingcombinationtreatmentwithantibioticand

antifungaleyedropsversusearlydiagnosisofthecauseandtargetedtherapy,

6. Ensurethatantifungaltreatments,especiallynatamycineyedrops,arereadily

accessibleeverywhere

7. Developprophylacticorpre-emptivetreatmentguidelinesforocularinjuries

8. Delivertraininginappropriatedeliveryoftheeyedrops,dosingandtiming

9. Encourage theperformanceof clinical trials todetermine thebest treatment

forfungalkeratitis.

JuanLuisRodriguezTudela

GAFFI,Geneva

February2021

References

1.Guidelinesforthemanagementofcornealulceratprimary,secondaryandtertiarycarehealthfacilitiesintheSouth-EastAsiaRegion.WHORegionalOfficeforSouth-EastAsia.Ophthal/126.

2.BadieeP.Mycotickeratitis,aStateoftheArtReview.JundishapurJMicrobiol2013;6:e8561.

3. Thomas PA, Kaliamurthy J. Mycotic keratitis: epidemiology, diagnosis andmanagement.ClinMicrobiolInfect.2013;19:210-20.

4. Leck AK, Thomas PA, Hagan M, et al. Aetiology of suppurative corneal ulcers inGhanaand south India, andepidemiologyof fungal keratitis.Br JOphthalmol. 2002;86:1211-5.

5. Brown L, LeckAK,GichangiG, BurtonMJ,DenningDW. The global incidence anddiagnosis of fungal keratitis. Lancet Infect Dis 2020. https://doi.org/10.1016/S1473-3099(20)30448-5

8

6.TilakR,SinghA,MauryaOP,etal.MycotickeratitisinIndia:afive-yearretrospectivestudy.JInfectDevCtries.2010;4:171-4.

7.BurtonMJ,PithuwaJ,OkelloE,etal.MicrobialkeratitisinEastAfrica:whyaretheoutcomessopoor?OphthalmicEpidemiol2011;18:158-63.

8. Stapleton F, Keay L, Edwards K, Naduvilath T, Dart JK, Brian G, Holden BA. Theincidence of contact lens-relatedmicrobial keratitis in Australia. Ophthalmol 2008;115:1655-62.

9.GovernmentofIndia.Nationalsurveyonblindness.1991-2001.Report2002.

10.WaddellK.ChildhoodblindnessandlowvisioninUganda.Eye1998;12:184–192.

11.DeorukhkarS,KatiyarR,SainiS.Epidemiologicalfeaturesandlaboratoryresultsofbacterial and fungal keratitis: a five-year study at a rural tertiary-care hospital inwesternMaharashtra,India.SingaporeMedJ2012;53:264-7.

12.JengBH,GritzDC,KumarAB,etal.EpidemiologyofulcerativekeratitisinNorthernCalifornia.ArchOphthalmol2010;128:1022-8.

13.Mselle J. Fungal keratitis as an indicator of HIV infection in Africa. Trop Doctor1999;29:133–35.

14. Thomas PA, Leck AK,MyattM. Characteristic clinical features as an aid to thediagnosisofsuppurativekeratitiscausedbyfilamentousfungi.BrJOphthalmol2005;89:1554-8.

15.OngHS,FungSS,MacleodD,DartJK,TuftSJ,BurtonMJ.Alteredpatternsoffungalkeratitis at a London ophthalmic referral hospital: an eight-year retrospectiveobservationalstudy.AmJOphthalmol2016;168:227–36.

16. Arunga S, Kintoki GM, Gichuhi S, et al. Delay along the care seeking journey ofpatientswithmicrobialkeratitisinUganda.OphthalmicEpidemiol2019;26:311–20.

17.KalavathyCM,ParmarP,Kaliamurthy J,etal.Comparisonof topical itraconazole1%withtopicalnatamycin5%forthetreatmentoffilamentousfungalkeratitis.Cornea2005;24:449-52.

18.PrajnaNV,KrishnanT,MascarenhasJ,etal.TheMycoticUlcerTreatmentTrial:Arandomized trial comparing natamycin vs voriconazole. Arch Ophthalmol 2012 Dec10:1-8.