Saudi Journal of Ophthalmology (2012) 26, 181–189

Ophthalmic Pathology Update

Endophthalmitis: A review of recent trends

Janice R. Safneck, MD, FRCPC ⇑

Abstract

Endophthalmitis is a feared complication of trauma, surgical procedures and septicemia. Although uncommon, its potential forsignificant visual loss is well recognized. Especially over the past decade, complicated surgeries and medical techniques haveincreased and seriously ill patients are being sustained in ever increasing numbers. New pathogens are being recognized andknown ones reclassified thanks to advances in molecular analysis. Continuously evolving PCR methodologies also add a newdimension to the diagnosis of infectious endophthalmitis. As well, medical literature is now truly international, encompassing stud-ies from around the world that expand our understanding of ocular infectious disease. This report reviews some of these changesas they relate to endophthalmitis and particularly to the spectrum of organisms involved.

Keywords: Endophthalmitis, Intraocular infection, Exogenous endophthalmitis, Endogenous endophthalmitis, Fungal endoph-thalmitis, Bacterial endophthalmitis

� 2012 Saudi Ophthalmological Society, King Saud University. All rights reserved.http://dx.doi.org/10.1016/j.sjopt.2012.02.011

Introduction

Endophthalmitis, defined as inflammation of one or morecoats of the eye and adjacent cavities, is an uncommonpotentially sight-threatening condition that varies geograph-ically in incidence and in cause. It may be categorized by clin-ical course (acute versus chronic), by etiology (infectiousversus non-infectious), by the route the causative agent en-ters the globe (exogenous versus endogenous) and by theorganism(s) involved (bacteria, fungi, parasites and rarely,viruses1). Certain organisms tend to be associated with par-ticular clinical settings, means of intraocular access and typesof inflammation (acute, chronic non-granulomatous, chronicgranulomatous or mixed cellular response).

Studies looking at the overall incidence of endophthalmitisyield differing results. For example, a recent British reportanalyzing acute endophthalmitis trends between 1991 and2004, found that of 120 cases, 59% were exogenous and41% endogenous in origin.2 In comparison, another series

Peer review under responsibilityof Saudi Ophthalmological Society,King Saud University

Received 29 January 2012; accepted 26 February 2012; available online 3 Marc

Departments of Pathology and Ophthalmology, University of Manitoba, Winn

⇑ Address: Department of Pathology, MS-459A, 820 Sherbrook St., Winnipeg1787; fax: +1 204 787 4942.e-mail address: safneck@cc.umanitoba.ca

from India of 955 patients presenting during a 10-year periodfound 92.6% exogenous endophthalmitis and only 7.4%endogenous endophthalmitis.3 Factors influencing variationbetween multiple reported studies include the period in timeanalyzed, prevalence of predisposing illnesses, geographicfactors in organism incidence, urban versus rural settings,ethnicity, outpatient versus inpatient populations especiallythose in tertiary care centers, and the small size of manyreported series.

Exogenous endophthalmitis

Exogenous endophthalmitis refers to infections resultingfrom breach of the globe exterior through surgery or trauma,or by fulminant progression of inflammatory processes suchas keratitis or scleritis.

Most postoperative endophthalmitis develops after cata-ract surgery as millions of these procedures are performedannually worldwide. Both acute endophthalmitis (generallyarising 24–48 h to a week post surgery) and chronic forms

j Production and hosting by ElsevierAccess this article online: www.saudiophthaljournal.comwww.sciencedirect.com

h 2012.

ipeg, Manitoba, Canada

, Manitoba, Canada R3A 1R9. Tel.: +1 204 787 1997, mobile: +1 204 791

182 J.R. Safneck

(presenting 4–6 weeks or more post surgery) may occur.Other procedures associated with varying risks of endoph-thalmitis include corneal surgeries (penetrating keratoplasty,keratoprosthesis insertion, refractive corneal surgeries), vitre-ous procedures (intravitreal injections, vitrectomies), glau-coma surgical treatments (blebs, glaucoma valveplacements), procedures to correct retinal detachmentincluding scleral buckling, and other miscellaneous ocularsurgeries, even strabismus correction. Most cases are spo-radic but occasionally are clustered, suggesting contami-nated materials/solutions or problems with instrumentsterilization likely are responsible.4,5 In these situations, unu-sual bacterial pathogens may be found.

Several recent studies from different parts of the world(India, China, Sweden, Norway, United Kingdom, USA) providepost cataract surgery endophthalmitis incidence figures rang-ing from 0.02% to 0.11% and post penetrating keratoplastyendophthalmitis from 0.108% to 0.5%.6–12 Studies from theBascom Palmer Institute in Florida covering 3 separate timeperiods (1984–1994, 1995–2001, 2002–2009) show a decreasein post cataract surgery endophthalmitis from 0.08% to 0.05%to 0.025% in the most recent series.10,11,13 The change wasattributed to progressive improvements in microsurgical andaseptic techniques. The effect of utilizing clear corneal insteadof scleral incisions has raised fears of increased infection;however, Lalwani et al.’s study14 also from Bascom Palmer,found similar clinical and microbiologic data although the meantime to presentation was increased.

Predominant symptoms include decreased vision and painwhile inability to visualize the fundus, pupillary fibrin membraneand hypopyon are the most common findings.15 Generallystudies have found coagulase negative Staphylococcus is thetop pathogen,6–8,14 comprising in some series more than 60%of isolates. Enterococci and streptococci were also frequentin a study from Sweden6. However, in 2 reports from India,the incidence of Gram positive and Gram negative bacteriawas approximately equal and more cases of Pseudomonasaeruginosa endophthalmitis were identified.9,16 Although lesscommon than bacterial cases, fungal endophthalmitis canoccur post cataract surgery,17,18 and is more frequent in Indianseries where detection of Aspergillus sp. eclipses Candida sp.18

The majority present acutely but latency periods of morethan 200 days have also been reported.18

In chronic endophthalmitis, a pale intracapsular plaque,fibrinous reaction in the anterior chamber and uveitis maybe seen. Histopathological examination of explanted intraoc-ular lens implants (IOLs) has revealed the presence of bacteriain relation to lens capsule and/or on the IOL surface, accom-panied by inflammatory cells including lymphocytes, plasmacells and neutrophils.19 Less commonly, fungi (especiallyCandida sp.) may be identified.20 Cultures, scanning and/ortransmission electron microscopy or DNA analysis21,22 havefound evidence of Staphylococcus sp. (especially S. epidermi-dis), Propionibacterium acnes, filamentous bacteria (includingActinomyces and Nocardia sp.) or mixed bacterial popula-tion.19,21,23–25 The bacteria are usually a part of conjunctivalflora (although occasional reported organisms such as Entero-coccus faecalis are not), and are capable of producing acuteendophthalmitis as well as chronic infections. They adhereto IOLs and create microcolonies through biofilm formationthat consists of microbes, glycocalyx and surface. Glycocalyxis a complex mixture of bacterial exopolysaccharides andlocally available cellular materials such as proteins and nucleic

acids that together act as a glue.26 In biofilm, organisms arewell protected from host inflammatory responses, bothphysically and through multiple genetic changes that alterantigenicity. However, as the organisms multiply, thosethat are shed outside the biofilm may revert to the originalform, rendering them vulnerable to attack. Organisms in thebiofilm are thus difficult to eradicate and may persist as anidus of infection despite apparent response to antibiotictreatments, resulting in relapsing endophthalmitis.22

Cornea-related endophthalmitis

The majority of endophthalmitis is bacterial27,28 andmainly due to Streptococcus sp. and Staphylococcus sp. Inone series, Gram negative organisms caused approximatelya quarter of infections.27 Fungal (Candida sp.) endophthalmi-tis also has been reported28–31 and may be increasing: caseshave been linked to donor-to-host spread and correlate withlonger donor cornea storage times.29 Recently, more limitedtransplantation such as Descemet stripping automated endo-thelial keratoplasty (DSAEK) is gaining favor. So far, therehave been limited reports of endophthalmitis post DSAEK,with Mycobacterium abscessus,32 Streptococcus pneumo-niae33 and Candida parapsilosis34 the pathogens involved.

In comparison, endophthalmitis secondary to keratitis ispredominantly bacterial, in both American and Indianseries.3,33,36 P. aeruginosa followed by Staphylococcus andStreptococcus sp. are common causes with filamentous fungiparticularly Aspergillus sp.36 and Fusarium sp.37 making up to40% of pathogens in one study.36 Contributing factors in-clude dry eye, corneal perforation, systemic immune dysfunc-tion and both topical and oral steroid use.35 Acanthamoebais well known for its ability to produce vision-threateningkeratitis that can be difficult to treat. However, it is highlyunusual for keratitis to progress to Acanthamoeba endoph-thalmitis38,39 even though in Kitzmann et al.’s series40 thisoccurred in 1 of 31 Acanthamoeba infected corneas.

Endophthalmitis associated with vitreous procedures

Both infectious and non-infectious endophthalmitis havebeen documented after intravitreal injections administeringtriamcinolone or anti-vascular endothelial growth factoragents (anti-VEGF). Risk is low but has increased as moreinjections are performed. Endophthalmitis incidence varieswidely. Deriving meaningful conclusions from published datarequires careful assessment since most series are retrospec-tive with non-uniform injection protocols of varying therapeu-tic agents. Some large recently published/cited examplesinclude McCannel’s meta-analysis of 105,531 anti-VEGFinjections with 54 instances of endophthalmitis (0.049%), 26of which were culture positive with Streptococcus sp. mostcommonly isolated (31%)41; Jager’s meta-analysis of all kindsof intravitreal injections with 24 cases of endophthalmitis per14,866 injections, most frequently due to Staphylococcussp.42; and Murray’s series of 34, 278 injections with 9 casesof endophthalmitis (0.03%), 5 of which were culture positive,predominantly Streptococcus.43 A subset of this latter groupperformed with standardized protocols, sterile techniquesand patient follow-up had just one case of endophthalmitis(Staphylococcus epidermidis) in 10,142 injections.43 Otherorganisms reported include Hemophilus influenzae and

Endophthalmitis: A review of recent trends 183

Mycobacterium chelonae abscessus.44 Differentiation be-tween infectious and non-infectious endophthalmitis is oftena challenge. Irigoyen et al.45 found commonly cited factorsfor distinction, i.e., lack of pain, conjunctival injection and ear-lier presentation, were not always helpful; in her study, theyapplied reasonably well to infectious endophthalmitis postanti-VEGF injection but were only rarely apparent in bacterialendophthalmitis following triamcinolone injections. Hergroup also noted 12 patients who had poor visual outcomesalthough this may have been at least in part due to underly-ing conditions. Since all endophthalmitis cases in the seriespresented within the first 48 h, vigilance from 24 to 72 h postinjection was recommended.

Concerns have been raised that increasing use of suture-less 25-gauge vitrectomy would lead to an increased inci-dence of exogenous endophthalmitis over that found with20-gauge pars plana vitrectomy. Study results have beenconflicting and a recent meta-analysis of 6 large series foundinterinstitutional differences in surgical practices and thepopulation base made drawing conclusion difficult.46 How-ever, diabetes mellitus is recognized as a risk factor for exog-enous endophthalmitis in all types of vitrectomy.

Glaucoma surgery endophthalmitis

Glaucoma related surgeries, blebs and implants (Molte-no, Baerveldt and Ahmed), also may be associated withinfectious endophthalmitis (Fig. 1). A large study of Ahmedglaucoma valves (542 eyes in 505 patients) from Riyadhfound endophthalmitis developed in 9 eyes (1.7%).47 Com-bining these findings with a review of reports from 1991to 2003 yielded 27 instances of endophthalmitis in 1427cases with an occurrence rate of 1.9%. Endophthalmitisdeveloped 63–330 days post valve insertion. Both acute

Figure 1. Mycobacterial endophthalmitis post Ahmed valve insertion (A). Ahematoxylin and eosin, �40). Acid fast bacilli are noted in suppurative inflam

and chronic forms occurred and children are much morelikely to be affected than adults. Conjunctival erosionoverlying the valve tube was common. Most frequent organ-isms were Haemophilus influenzae and Streptococcus sp.and less commonly, P. aeruginosa.47 Similarly, both acuteand delayed onset endophthalmitis may develop with filter-ing blebs, reported to occur in 0.2–9.6% of cases. In a re-cent series of 71 eyes from 68 patients, one quarter ofpatients had a bleb leak and 83% were culture positive, pre-dominantly for Streptococcus sp., a variety of Gram nega-tive organisms (Moraxella sp., P. aeruginosa, H. influenzae,Serratia sp.) and coagulase negative Staphylococcus. Thir-teen percent (9 eyes) had polymicrobial infections and thesame percentage ultimately was eviscerated/enucleated.48

More virulent organisms than seen in post cataract surgeryendophthalmitis and poor visual outcomes have been notedin other studies as well.48

Exogenous endophthalmitis secondary to trauma

Posttraumatic endophthalmitis has an overall frequency of<1–18.9% in recently reported series.49–54 Intraocular cultureresults post-trauma do not necessarily correlate with thepresence of endophthalmitis.53 Onset may be acute or de-layed but the most virulent organisms can destroy an eye inhours. Risk factors for infection differ in various series andcomparison can be challenging as series are usually retro-spective and treatment practices not consistent; outcomesalso vary widely with trauma and endophthalmitis relateddamage together impairing successful recovery. Amongstthe best results reported were visual acuity of 20/200 or bet-ter in 67% of patients55 Longer time between injury andexamination, poorer visual acuity at presentation, virulenceof organisms, and the presence of an intraocular foreign

bundant inflammation was seen especially at the tube insertion site (B,mation without giant cells or granulomas (C, ZN �600).

184 J.R. Safneck

body (IOFB) were often reported as significant for endoph-thalmitis development.50,53,56,57

In both adult and pediatric series of posttraumaticendophthalmitis from around the world, Staphylococcusand Streptococcus sp. are the most frequent pathogens55,58,59 P. aeruginosa and B. cereus also are frequently identi-fied51,60 but their prevalence varies with geographic location.

Bacillus sp. infections are noteworthy for the rapiddestruction they can cause, with visual loss and often lossof the eye in hours to a few days after injury (Fig. 2). In tworetrospective series, only 15 of 31 and 8 of 22 patients withBacillus sp. endophthalmitis ultimately had visual acuity bet-ter than 20/400.61,63 Bacillus sp. have been reported asresponsible for 15–46% of post-traumatic endophthalmitisbut are only rarely isolated from post-surgical endophthalmi-tis cases.62 B. cereus is most commonly found in America62

while in an Indian study,61 a variety of species were isolatedand some infections were caused by a mixture of Bacillussp. Bacillus organisms are spore-forming Gram positive rods,ubiquitous in soil, water and dust. Initial reports suggestedthat these bacteria were particularly associated with metalobject lacerations but a variety of trauma causes have beenreported. Virulence is attributable to bacterial toxins – hem-olysins, lipases, enterotoxins and proteases – acting to-gether.63 As well, wild-type motile strains are more virulentthan non-motile ones as they can cause not only posteriordisease but also anterior segment destruction.64 Experimen-tal studies have demonstrated that decline in retinal functionand neutrophilic infiltration of vitreous can occur as fast as 4 hpost-infection,65 suggesting that therapeutic interventionsmust be initiated rapidly if vision is to be saved.

In two recent studies from Saudi Arabia56 and India,66 fun-gi made up 3.8% and 7.3% of post-traumatic endophthalmitisrespectively. Gupta et al noted filamentous fungi especiallyAspergillus sp. and Fusarium sp. were usually isolated.66

The incidence of endophthalmitis associated with IOFBsvaries but they generally increase the risk of endophthalmitisover that associated with open globe injury alone. Bhagatet al.57 in their extensive literature review of post-traumaticendophthalmitis noted a range of 3.1–30% without an IOFB

Figure 2. Bacillus sp. post-traumatic endophthalmitis (A) leading to evisceratiwere seen in the inflammatory debris (B, Gram stain, �1000).

and 1.3–60% with an IOFB.57 Degree of risk is dependenton size and composition of the foreign body including anycontaminating materials such as soil, the speed with whichit enters the globe, its path within the eye, the length of timebetween injury and foreign body removal, and the immunesystem of the affected individual in addition to treatmentundertaken. Most endophthalmitis associated with IOFBs isbacterial, and Bacillus sp. are particularly of concern in thissetting. In addition, vision-threatening sterile reactiveendophthalmitis also can develop from IOFBs, in particularwith metallic IOFBs, especially copper.57

Endogenous endophthalmitis

Endogenous or metastatic endophthalmitis is typically theresult of hematogenous spread. It is highly unusual for indi-viduals with endogenous endophthalmitis to have no risk fac-tors for systemic infection67 although it has been reported.68

Patients generally have a history of chronic illness (diabetes,HIV, malignancy, intravenous drug use), transplantation,immunosuppressive therapy, and/or catheterization. Bacteriaand fungi are the most common pathogens, the former typ-ically Staphylococcus sp., Streptococcus sp. and Klebsiellapneumoniae, the latter generally Candida sp. or Aspergillussp. A variety of organisms, including agents that are littleknown and/or otherwise of limited virulence have beenincreasingly found.69 Even protozoa such as Microsporidia70

and amoebae71 may be pathogens, including in patients withno history of HIV infection. In culture proven endogenousendophthalmitis, fungi typically exceed bacteria as iso-lates,72–74 62% versus 38% as illustrated by one recentstudy75 but in some but not all East Asian populations, bacte-ria are more frequent.76

The vast majority of affected individuals are adults withpediatric endogenous endophthalmitis representing be-tween 0.1% and 4% of cases, incidence dependent on cohortlocation with higher rates in India than in America.77 Patientsmay have overt systemic sepsis, or the initiating focus ofinfection may be occult, consisting of either multiorgan

on 48 h after penetrating injury with metallic shards. Numerous organisms

Endophthalmitis: A review of recent trends 185

subclinical involvement or a localized clandestine nidus ofpathogens, possibly with only transient bacteremia/fungemiaas an explanation for the infective endophthalmitis.78,79 Boththese scenarios can present diagnostic perils. Patients with-out obvious systemic infection may have their endophthalmi-tis misdiagnosed,75,80,81 or have it thought to be a strictlyocular process, leaving unrecognized possibly devastatingdisease elsewhere (for example, endocarditis). Seriously illpatients may neglect or be unable to mention eye symptomsand medical staff may be so focused on other pressing issuesthat endophthalmitis is overlooked until vision is permanentlycompromised.

Some literature reports suggest endogenous endophthal-mitis is increasing, perhaps because of expanded transplan-tation, more complicated surgeries and the ability to keepserious ill patients alive longer. Consequently, it has beenrecommended that ophthalmic screening should be routinein high risk situations such as intravenous drug use, long termantibiotics, immunosuppressive therapy, primary or second-ary immunodeficiency, prolonged central line use, debilitatedpatients and premature infants.82 On the other hand, even inthe setting of culture-proven sepsis, endogenous endoph-thalmitis is unusual. Lingappan et al.83 found no cases ofendophthalmitis in the routine screening of patients withfungemia. Feman et al.84 observed that of 82 patients withsystemic fungal disease, only 2 patients developed endoph-thalmitis and Dozier et al.85 noted that of 211 patients withpositive fungal cultures at a tertiary care center, less than1% had ocular involvement.

These apparently conflicting findings illustrate that endog-enous endophthalmitis is a complex subject. In part, this re-lates to the patient population studied, their risk factors, tothe organism(s) involved and to the early administration ofantibiotic/antifungal agents in bacteremic/fungemic patients.

Candida sp. are the most commonly reported causes ofendogenous endophthalmitis,67,86,87 perhaps not surprisinglyas yeasts including Candida make up 25% of blood streaminfection acquired in American hospitals.88 However, whilehistorically the rate of intraocular Candida sp. infectionswas between 9% and 45%,67 current rates are around 1–2%likely due to the early identification of Candida and promptadministration of antifungal agents in septic patients.84,89

Two recent series are illustrative of this point. In a Scandina-

Figure 3. Bilateral Aspergillus sp. endophthalmitis in a renal transplant patien(A). Globe at autopsy with hazy vitreous and pale thickened retina (B). Microswith fungal hyphae (C, hematoxylin and eosin, �200).

vian study of 203 patients with Candidemia, 9 cases of retini-tis and 1 of endophthalmitis were detected.88 In a similarAmerican study of 38 patients with Candidemia, 8 had chori-oretinitis but none had endophthalmitis.90 In excess of 150Candida species are documented human pathogens, yetCandida albicans remains the most important although oth-ers such as C. glabrata, C. tropicalis, C. dubliniensis and C.krusei are being increasingly detected.91 C. albicans has beenshown in several reports to have a greater risk of progressionfrom chorioretinitis to endophthalmitis, in keeping withgreater virulence and prevalence,89,90 while the opposite istrue of C. parapsilosis.89 Risk factors for the developmentof Candida endogenous fungal endophthalmitis include dia-betes, antibiotic use, gastrointestinal surgery, catheters, solidtumors, intravenous drug use and immunosuppres-sion.78,79,87,92–94 Many patients with identified ocular diseaseare without ocular symptoms89 or present in a subacute fash-ion with altered vision, low-grade pain, photophobia andinjection.90 Experimental models of C. albicans endogenousendophthalmitis have shown the initial lesions appear at theequator then spread posteriorly, and animals with prior ste-roid treatment had more fungal pseudohyphae than thosewithout.95 Histopathologic study of clinical enucleation spec-imens have demonstrated Candida organisms in vitreous withassociated suppurative non-granulomatous inflammation andtypically minimal if any choroidal or retinal necrosis.93 Thesefindings are consistent with 76–85% of endogenous Candidaendophthalmitis patients having final visual acuities of 20/200or better.78,96

In studies of endogenous fungal endophthalmitis, Asper-gillus sp. have been reported less frequently than Candidasp., 75% versus 25% in an American study83 but are muchmore visually devastating.75,79,97,98 Aspergillus endogenousendophthalmitis often presents with nonspecific symptomsbut rapidly progressive retinal damage soon occurs (Fig. 3).Histopathological analysis of enucleated globes has showninvolvement of retinal and choroidal vasculature with retinalnecrosis and predominantly a neutrophilic inflammatory re-sponse although chronic granulomatous inflammation maybe found.93,98 Septate fungi with 45� angle branching aretypical of Aspergillus sp. but do not permit definitive diagno-sis as some other molds (e.g. Monosporium apiospermum/Pseudoallescheria boydii) can be morphologically similar.

t. Endocarditis was suspected but not confirmed until just prior to deathcopy revealed abundant neutrophilic inflammation in vitreous and retina

186 J.R. Safneck

Culture and/or PCR technique are needed for definitive diag-nosis. Aspergillus fumigatus is the most common speciesfound with A. flavus second. However, in some studies, Fusar-ium sp. outnumber Aspergillus sp.79

Risk factors for endogenous Aspergillus endophthalmitisinclude corticosteroid use, intravenous drug use, solid organand stem cell transplantation and lung disease, and many pa-tients have more than one underlying contributing condi-tion.69,79,98 Approximately three quarters of cases areunilateral and one quarter ultimately involve both eyes. Morethan half have spread to other organs. Blurred vision, eyepain and redness are common and in one study, vitritis wasobserved in 45 of 73 cases and retinal lesions or chorioretini-tis in another 28. Only 2 patients had anterior chamberinvolvement.98 This distribution is consonant with vitrectomyas the most reliable means of organism diagnosis.79,83

Recovery of useful vision has been reported in 0–10%.75,98

Just over half of the patients have a diagnosis of Aspergillusendophthalmitis made only after autopsy98 and in cancer pa-tients, filamentous fungal endophthalmitis such as Aspergil-lus sp. is a marker for high mortality.69,79 Brain, lung andheart (endocarditis) are other typical sites of involvement atpost-mortem.

Endogenous bacterial endophthalmitis is much less fre-quent than endogenous fungal endophthalmitis or exoge-nous bacterial endophthalmitis with an incidence of 2–11%.80 Age ranged from neonates to patients greater than90 years old with a mean of 50–60 years of age dependingon the population/literature assessed.80 Nineteen percentwere bilateral in one study.99 Patients often complain ofblurred vision with ocular pain and most but not all have evi-dence of systemic illness. Examination reveals hypopyon andhazy ocular media.80,99 Both Gram positive and Gram nega-tive bacteria have been reported as pathogens with inci-dences differing around the world. For example, severalreports of East Asian populations76,99–101 have clearly docu-mented K. pneumoniae as the predominant endogenousendophthalmitis cause, with hepatobiliary infections (espe-cially liver abscesses), as the initial disease site and diabetesmellitus a common co-existing condition.99 Elsewhere in theworld, S. aureus, Group B Streptococcus and S. pneumoniaehave been more frequent.68,80,81 In Caucasians, endocarditis,skin/joint and urinary tract infections were common organismsources.80,87,99,102 Other noteworthy pathogens include P.aeruginosa which is more frequent in individuals youngerthan 25 years of age including neonates80 and Neisseria men-ingitidis which was noted to be decreasing in incidence in the1980s but still a factor in childhood infections.103 Escherichiacoli has been isolated from endogenous endophthalmitis pa-tients with urinary tract infections80 and from seeminglyhealthy individuals who have endogenous bacterial endoph-thalmitis.68 B. cereus has been associated with intravenousdrug use and, as in trauma-related cases, commonly pro-gressed to evisceration/enucleation.80,103

As with endogenous fungal endophthalmitis, visual out-comes are often poor regardless of the organism in-volved.68,80,99,104 In Wong’s East Asian study, good visualacuity returned in only 28% of cases.99 Similarly, Jackson inhis British cohort found 81% of affected eyes with light per-ception or worse.80 More recent cases of K. pneumoniaehave had more patients with visual acuities of at least 10/200, perhaps due to more clinical experience with the dis-ease. However, overall results are still unsatisfactory with

approximately three-quarters of patients having a visual acu-ity of 1.101 Statistical analysis has pointed to the virulence ofK. pneumoniae as the most important factor.99 This organismhas been isolated from multiple types of clinical specimensand may colonize up to 20% of hospitalized patients.100 Ani-mal models of K. pneumoniae endophthalmitis have shownhypermucoviscosity phenotypes, K1 and K2 capsular pheno-types and aerobactin production as virulence indicators. Re-cently strains that produce beta-lactamase enzymesconferring resistance to antibiotics frequently employed inthe treatment of Gram negative endophthalmitis have beendiscovered.100 Experimentally, retinal destruction has ap-peared as soon as 48 h post infection,100 so patients withposterior infections and cloudy media at presentation aremore likely to do poorly.99 Timely diagnosis and treatmentare essential.

Diagnosis

Aspirate of aqueous and/or vitreous for Gram stain is com-monly employed for diagnosis; although helpful when posi-tive for organisms, it is often not very sensitive. In the past,culture has been the only available means of establishingorganism(s) causing endophthalmitis. As previously men-tioned, in endogenous endophthalmitis, vitrectomy cultureshave proven the most reliable with poor concordance notedbetween anterior chamber cultures and vitreous isolates.105

Recently, PCR has been found to be an important diagnostictechnique and has been reported to greatly aid in bacterialand fungal detection.106,107 For example, in one real-timePCR study, detection of bacteria improved from 47.6% to95.3%.107 As with cultures, specimen contamination will yieldfalse positive results and poorly stored specimens can be sodegraded that they are not helpful. Another advantage ofPCR is its rapidity: results can be ready in as little as90 min,87 a tremendous advantage over cultures, especiallyfor some slow-growing organisms such as mycobacteria. Fur-ther, it can be performed on fixed paraffin embedded histol-ogy specimens108,109 to identify organisms not diagnosedpresurgery/autopsy. PCR techniques are important diagnos-tic tools that hopefully will become more widely available.

Non-infectious endophthalmitis

This condition has diverse etiologies, and includes sys-temic autoimmune diseases, local ocular inflammations of un-known cause, endophthalmitis related to lens material andendophthalmitis attributable to intraocular foreign bodies.

Phacoanaphylactic endophthalmitis (lens induced granu-lomatous inflammation) is a rare consequence of lens injuryand represents an autoimmune response to lens protein towhich the body is normally tolerant.110 This may occur aftertrauma, as a result of inflammation from other causes ruptur-ing the lens capsule, or post surgery such as following extra-capsular cataract extraction when residual lens cortex ispresent. Histopathologically, a mixed neutrophilic and granu-lomatous response is centered around the lens, typically in azonal pattern: neutrophils associated with degenerating lensmaterial are surrounded by giant cells and histiocytes whichare in turn encircled by chronic non-granulomatous inflamma-tion. The condition is suspected clinically in only 5% of histo-logically proven cases and thus is usually a surprise at

Endophthalmitis: A review of recent trends 187

enucleation. However, diagnosis is also possible in vivo onfine needle aspiration biopsy.111 Seventy percent of casesare associated with patchy choroidal non-granulomatousinflammation, 60% with retinal detachment, 50% with retinalmononuclear cell perivasculitis, 30% with optic atrophy and3–7% with sympathetic ophthalmia.110,112

The term ‘‘phacotoxic’’ endophthalmitis was previouslyused to cover a mixed group of conditions related to intraoc-ular lens implant surgery. As the inflammation is centered ante-riorly, the term toxic anterior segment syndrome (TASS) ismore appropriate. Causes include reactions to chemicals (irri-gation solutions, preservatives, drugs, denatured viscoelas-tics), IOL materials, instrument sterilization and preparation-related compounds and bacterial endotoxins.113,114 The latterare the result of bacterial growth killed by product sterilizationwith the heat-resistant endotoxin remaining. In sufficient lev-els, the endotoxin may produce sterile endophthalmitis.115

Sterile endophthalmitis has been noted as well following intra-ocular injection, post vitrectomy and after surgery includingglaucoma drainage device implantation. Similar to TASS, thecondition may be multifactorial. Intravitreal triamcinolone ace-tonide has been associated with pseudoendophthalmitiscaused by triamcinolone crystals migrating into the anteriorchamber; infectious and sterile endophthalmitis have alsobeen reported in association with this chemical.116

Clinically sterile endophthalmitis and TASS are differenti-ated from infectious endophthalmitis by the former entitiestypically arising within 24 h of surgery, being Gram stainand culture negative, involving the anterior segment in thecase of TASS, and showing no effect from antibiotics butimprovement with topical and/or oral steroids. However, dis-tinction may be problematic as some infectious cases mayhave rapid onset and have initial negative cultures.113

Conclusion

Through an improved understanding of the causes andepidemiology of endophthalmitis will come further measuresfor decreasing the incidence of this dreaded disease. Wide-spread use of molecular techniques such as real-time PCRpromise to lessen diagnostic times and provide more accu-rate diagnosis, permitting faster targeted therapies. On theother hand, the rise of organisms such as methicillin resistantS. aureus and other ‘‘super bugs’’ will necessitate constantvigilance so that the successes achieved so far will not be lost.

Acknowledgements

The author wishes to thank Dr. Lisa Gould and Dr. StephenBrodovsky for clinical information pertaining to the case illus-trated in Fig. 1.

References

1. Borne MJ, Shields JA, Shields CL, De Porter P, Ehya H. Bilateral viralendophthalmitis as the presenting sign of severe combinedimmunodeficiency. Arch Ophthalmol 1994;112:1280–1.

2. Krause L, Bechrakis N, Heimann H, Kildal D, Foerster M. Incidenceand outcome of endophthalmitis over a 13-year period. Can JOphthalmol 2009;44:88–94.

3. Ramakrishnan R, Bharathi M, Shivkumar C, et al. Microbiologicalprofile of culture-proven cases of exogenous and endogenous

endophthalmitis: a ten-year retrospective study. Eye (Lond)2009;23:945–56.

4. Akçakaya AA, Sargin F, Erbil HH, et al. A cluster of acute-onsetpostoperative endophthalmitis over a 1-month period: investigationof an outbreak caused by an uncommon species. Br J Ophthalmol2011;95:481–4.

5. Korah S, Braganza A, Jacob P, Balaji V. An ‘‘epidemic’’ of postcataract surgery endophthalmitis by a new organism. Indian JOphthalmol 2007;55:464–6.

6. Wejde G, Montan P, Lundstrom M, Stenevi U, Thorburn W.Endophthalmitis following cataract surgery in Sweden: nationalprospective survey 1999–2001. Acta Ophthalmol Scand 2005;83:7–10.

7. Mollan SP, Gao A, Lockwood A, Durrani OM, Butler L. Postcataractendophthalmitis: incidence and microbial isolates in a UnitedKingdom region from 1996 through 2004. J Cataract Refract Surg2007;33:759.

8. Cao X, Liu A, Zhang J, et al. Clinical analysis of endophthalmitisafter phacoemulsification. Can J Ophthalmol 2007;42:844–8.

9. Jamulingam M, Parameswaran SK, Lysa S, Selvaraj M, MadhavanHN. A study on the incidence, microbiological analysis andinvestigations on the source of infection of postoperativeinfectious endophthalmitis in a tertiary care ophthalmic hospital:an 8-year study. Indian J Ophthalmol 2010;58:297–302.

10. Eifrig CWG, Flynn Jr HW, Scott IU, Newton J. Acute-onsetpostoperative endophthalmitis: review of incidence and visualoutcomes (1995–2001).Erratum inOphthalmic SurgLasers 2003;34:80.

11. Wycoff CC, Parrott MB, Flynn Jr HW, Shi W, Miller D, Alfonso EC.Nosocomial acute-onset postoperative endophthalmitis at auniversity teaching hospital (2002–2009). Am J Ophthalmol 2010;150:392–8.

12. Sandvig KU, Dannevig L. Postoperative endophthalmitis:establishment and results of a national registry. J Cataract RefractSurg 2003;29:1273–80.

13. Aaberg Jr TM, Flynn Jr HW, Schiffman J, Newton J. Nosocomialacute-onset postoperative endophthalmitis survey. Ophthalmology1998;105:1004–10.

14. Lalwani GA, Flynn Jr HW, Scott IU, et al. Acute-onsetendophthalmitis after clear corneal cataract surgery (1996–2005).Clinical features, causative organisms, and visual acuity outcomes.Ophthalmology 2008;115:473–6.

15. Cornut PL, Thuret G, Creuzot-Garcher C, et al. Relationshipbetween baseline clinical data and microbiologic spectrum in onehundred patients with acute postcataract endophthalmitis. Retina2012;32:549–57.

16. Anand AR, Therese KL, Madhavan HN. Spectrum of aetiologicalagents of postoperative endophthalmitis and antibiotic susceptibilityof bacterial isolates. Indian J Ophthalmol 2000;48:123–8.

17. Gregori NZ, Flynn Jr HW, Miller D, et al. Clinical features,management strategies, and visual acuity outcomes of Candidaendophthalmitis following cataract surgery. Ophthalmic Surg LasersImaging 2007;38:378–85.

18. Narang S, Gupta A, Gupta V, et al. Fungal endophthalmitisfollowing cataract surgery: clinical presentation, microbiologicalspectrum, and outcome. Am J Ophthalmol 2001;132:609–17.

19. Adán A, Casaroli-Marano RP, Gris O, et al. Pathological findings in thelens capsules and intraocular lens in chronic pseudophakic endoph-thalmitis: an electron microscopy study. Eye (Lond) 2008;22:113–9.

20. Rensch F, Schlichtenbrede FC, Jonas JB. Postoperative mycoticendophthalmitis. J Cataract Refract Surg 2010;36:1233–4.

21. Lai J-Y, Chen K-H, Lin Y-C, Hsu W-M, Lee S-M. Propionibacteriumacnes DNA from an explanted intraocular lens detected bypolymerase chain reaction in a case of chronic pseudophakicendophthalmitis. J Cataract Refract Surg 2006;32:522–5.

22. Miller KV, Eisley KM, Shanks RM, et al. Recurrent enterococcalendophthalmitis seeded by an intraocular lens biofilm. J CataractRefract Surg 2011;37:1355–9.

23. Wenkel H, Rummelt V, Knorr H, Naumann GO. Chronicpostoperative endophthalmitis following cataract extraction andintraocular lens implantation. Report on nine patients. Ger JOphthalmol 1993;2:419–25.

24. Samson CM, Forster CS. Chronic postoperative endophthalmitis. IntOphthalmol Clin 2000;40:57–67.

25. Al-Mezaine HS, Al-Assiri A, Al-Rajhi AA. Incidence, clinical features,causative organisms, and visual outcomes of delayed-onsetpseudophakic endophthalmitis. Eur J Ophthalmol 2009;19:804–11.

188 J.R. Safneck

26. Kodjikian L, Burillon C, Roques C, et al. Intraocular lenses, bacterialadhesion and endophthalmitis prevention: a review. Biomed MaterEng 2004;14:395–409.

27. Kunimoto DY, Tasman W, Rapuano C, et al. Endophthalmitis afterpenetrating keratoplasty: microbiologic spectrum and susceptibilityof isolates. Am J Ophthalmol 2004;137:343–5.

28. Hassan SS, Wilhelmus KR. Medical Review Subcommittee of the EyeBank Association of America. Eye-banking risk factors for fungalendophthalmitis compared with bacterial endophthalmitis aftercorneal transplantation. Am J Ophthalmol 2005;139:685–90.

29. Hassan SS, Wilhelmus KR, Dahl P, et al. Infectious disease riskfactors of corneal graft donors. Arch Ophthalmol 2008;126:235–9.

30. Keyhani K, Seedor JA, Shah MK, Terraciano AJ, Ritterband DC. Theincidence of fungal keratitis and endophthalmitis followingpenetrating keratoplasty. Cornea 2005;24:288–91.

31. Schotveld JH, Raijmakers AJ, Henry Y, Zaal MJ. Donor-to-hosttransmitted Candida endophthalmitis after penetratingkeratoplasty. Cornea 2005;24:887–9.

32. Chang V, Karp CL, Yoo SH, et al. Mycobacterium abscessusendophthalmitis after Descemet’s stripping with automatedendothelial keratoplasty. Cornea 2010;29:586–9.

33. Shih CY, Ritterband DC, Rubino S, et al. Visually significant andnonsignificant complications arising from Descemet strippingautomated endothelial keratoplasty. Am J Ophthalmol 2009;148:837–43.

34. Chew AC, Mehta JS, Li L, Busmanis I, Tan DT. Fungalendophthalmitis after descemet stripping automated endothelialkeratoplasty – a case report. Cornea 2010;29:346–9.

35. Scott IU, Flynn Jr HW, Feuer W, et al. Endophthalmitis associatedwith microbial keratitis. Ophthalmology 1996;103:1864–70.

36. Bharathi MJ, Ramakrishnan R, Shivakumar C, Meenakshi R, LionalrajD. Etiology and antibacterial susceptibility pattern of community-acquired bacterial ocular infections in a tertiary eye care hospital insouth India. Indian J Ophthalmol 2012;58:497–550.

37. Shen Y-C, Wang C-Y, Tsai H-Y, Lee H-N. Intracameral voriconazoleinjection in the treatment of fungal endophthalmitis resulting fromkeratitis. Am J Ophthalmol 2010;149:916–21.

38. Davis MJ, Packo KH, Epstein RJ, Grostern RJ, Cohen JA.Acanthamoeba endophthalmitis following penetrating keratoplastyfor Acanthamoeba keratitis. Arch Ophthalmol 2010;128:505–6.

39. Clarke DW, Niederkorn JY. The pathophysiology of Acanthamoebakeratitis. Trends Parasitol 2006;22:175–80.

40. Kitzmann AS, Goins KM, Sutphin JE, Wagoner MD. Keratoplasty fortreatment of Acanthamoeba keratitis. Ophthalmology 2009;116:864–9.

41. Moshfeghi AA. Endophthalmitis following intravitreal anti-vascularendothelial growth factor injections for neovascular age-relatedmacular degeneration. Semi Ophthalmol 2011;26:139–48.

42. Jager RD, Aiello LP, Patel SC, Cunningham Jr ET. Risks ofintravitreous injection: a comprehensive review. Retina 2004;24:676–98.

43. Cavalcante LL, Cavalcante ML, Murray TG, et al. Intravitrealinjection analysis at the Bascom Palmer Eye Institute: evaluation ofclinical indications for the treatment and incidence rates ofendophthalmitis. Clin Ophthalmol 2010;4:519–24.

44. Schwartz SG, Flynn HW, Scott IU. Endophthalmitis after intravitrealinjections. Expert Opin Pharmacother 2009;10:2119–26.

45. Irigoyen C, Ziahosseini K, Morphis G, Stappler T, Heimann H.Endophthalmitis following intravitreal injections. Graefes Arch ClinExp Ophthalmol 2011; published, online 03 November 2010.

46. Bahrani HM, Fazelat AA, Thomas M, et al. Endophthalmitis in theera of small gauge transconjunctival sutureless vitrectomy – metaanalysis and review of the literature. Semi Ophthalmol 2010;25:275–82.

47. Al-Torbak AA, Al-Shahwan A, Al-Jadaan I, Al-Hommadi A, EdwardDP. Endophthalmitis associated with the Ahmed glaucoma valveimplant. Br J Ophthalmol 2005;89:454–8.

48. Leng T, Miller D, Flynn Jr HW, Jacobs DJ, Gedde SJ. Delayed-onsetbleb-associated endophthalmitis (1996–2008). Causative organismsand visual acuity outcomes. Retina 2011;31:344–52.

49. Fan JC, Niederer RL, von Lany H, Polkinghorne PJ. Infectiousendophthalmitis: clinical features, management and visualoutcomes. Clin Experiment Ophthalmol 2008;36:631–6.

50. Andreoli CM, Andreoli MT, Kloek CE, Ahuero AE, Vavvas D, DurandML. Low rate of endophthalmitis in a large series of open globeinjuries. Am J Ophthalmol 2009;147:601–8.

51. Chhabra S, Kunimoto DY, Kai L, et al. Endophthalmitis after openglobe injury: Microbiologic spectrum and susceptibilities of isolates.Am J Ophthalmol 2006;142:852–4.

52. Faghihi H, Hajizadeh F, Esfahani MR, et al. PosttraumaticEndophthalmitis. Report No. 2. Retina 2012;32:146–51.

53. Essex RW, Yi Q, Charles PG, Allen PJ. Post-traumaticendophthalmitis. Ophthalmology 2004;111:2015–22.

54. Zhang Y, Zhang MN, Jiang CH, Yao Y, Zhang K. Endophthalmitisfollowing open globe injury. Br J Ophthalmol 2010;94:111–4.

55. Alfaro DV, Roth DB, Laughlin RM, Goyal M, Liggett PE. Paediatricpost-traumatic endophthalmitis. Br J Ophthalmol 1995;79:888–91.

56. Al-Omran AM, Abboud EB, Abu El-Asrar AM. Microbiologicspectrum and visual outcome of posttraumatic endophthalmitis.Retina 2007;27:236–42.

57. Bhagat N, Nagori S, Zarbin M. Post-traumatic infectiousendophthalmitis. Surv Ophthalmol 2011;56:214–51.

58. Al-Rashaed SA, Abu El-Asrar AM. Exogenous endophthalmitis inpediatric age group. Ocular Immunol Inflam 2006;14:285–92.

59. Wykoff CC, Flynn Jr HW, Miller D, Scott IU, Alfonso EC. Exogenousfungal endophthalmitis; microbiology and clinical outcomes.Ophthalmology 2008;115:1501–7.

60. Vedantham V, Nirmalan PK, Ramasamy K, Prakash K,Namperumalsamy P. Clinico-microbiological profile and visualoutcomes of post-traumatic endophthalmitis at a tertiary eye carecentre in South India. Indian J Ophthalmol 2006;54:5–10.

61. Das T, Choudhury K, Sharma S, Jalali S, Nuthethi RtheEndophthalmitis Research Group. Clinical profile and outcome inBacillus endophthalmitis. Ophthalmology 2001;108:1819–25.

62. Miller JJ, Scott IU, Flynn Jr HW, et al. Endophthalmitis caused byBacillus species. Am J Ophthalmol 2008;145:883–8.

63. Callegan MC, Kane ST, Cochran DC, et al. Bacillus endophthalmitis:roles of bacterial toxins and motility during infection. InvestOphthalmol Vis Sci 2005;46:3233–8.

64. Callegan MC, Novosad BD, Ramiez R, Ghelardi E, Senesi S. Role ofswarming migration in the pathogenesis of bacillus endophthalmitis.Invest Ophthalmol Vis Sci 2006;47:4461–7.

65. Ramadan RT, Ramirez R, Novosad BD, Callegan MC. Acuteinflammation and loss of retinal architecture and function duringexperimental Bacillus endophthalmitis. Curr Eye Res2006;31:955–65.

66. Gupta A, Srinivasan R, Kaliaperumal S, Saha I. Post-traumatic fungalendophthalmitis—a prospective study. Eye (Lond) 2008;22:13–7.

67. Takebayashi H, Mizota A, Tanaka M. Relation between stage ofendogenous fungal endophthalmitis and prognosis. Graefe ArchClin Exp Ophthalmol 2006;244:816–20.

68. Shankar K, Gyanendra L, Hari S, Narayan SD. Culture provenendogenous bacterial endophthalmitis in apparently healthyindividuals. Ocul Immunol Inflamm 2009;17(6):396–9.

69. Vergoulidou M, Krause L, Foerster MH, Thiel E, Schwartz S.Endogenous filamentous fungal endophthalmitis – single-centresurvey in patients with acute leukaemia or postallogeneic stem celltransplantation and review of the literature. Mycoses 2011;54:e704–11.

70. Yoken J, Forbes B, Maguire AM, Prenner JL, Carpentieri D.Microsporidial endophthalmitis in a patient with acutemyelogenous leukemia. Retina 2002;22:123–5.

71. Matsuo T, Notohara K, Shigara F, Yumiyama S. Endogenousamoebic endophthalmitis. Arch Ophthalmol 2001;119:125–8.

72. Leibovitch I, Lai T, Raymond G, Zadeh R, Nathan F, Selva D.Endogenous endophthalmitis: a 13-year review at a tertiary hospitalin South Australia. Scand J Infect Dis 2005;37:184–9.

73. Ness T, Pelz K, Hansen LL. Endogenous endophthalmitis:microorganisms, disposition and prognosis. Acta OphthalmolScand 2007;85:852–6.

74. Zhang H, Liu Z. Endogenous endophthalmitis: a 10-year review ofculture-positive cases in northern China. Ocul Immunol Inflamm2010;18:133–8.

75. Schiedler V, Scott I, Flynn Jr HW, et al. Culture-proven endogenousendophthalmitis: clinical features and visual acuity outcomes. Am JOphthalmol 2004;137:725–31.

76. Chung K, Kim Y, Song Y, et al. Clinical review of endogenousendophthalmitis in Korea: a 14 year review of culture positive casesof two large hospitals. Yonsei Med J 2011;52:630–4.

77. Chaudhry IA, Shamsi FA, Al-Dhibi H, Khan AO. Pediatricendogenous bacterial endophthalmitis: case report and review ofthe literature. J AAPOS 2006;10:491–3.

Endophthalmitis: A review of recent trends 189

78. Binder MI, Chua J, Kaiser PK, Procop GW, Isada CM. Endogenousendophthalmitis. An 18-year review of culture-positive cases at atertiary care center. Medicine (Baltimore) 2003;82:97–105.

79. Lamaris GA, Esmaeli B, Chamilos G, et al. Fungal endophthalmitis ina tertiary care cancer center: a review of 23 cases. Eur J ClinMicrobial Infect Dis 2008;27:343–7.

80. Jackson TL, Eykyn SJ, Graham EM, Stanford MR. Endogenousbacterial endophthalmitis: a 17-year prospective series and reviewof 267 reported cases. Surv Ophthalmol 2003;48:403–23.

81. Okada AA, Johnson RP, Liles WC, D’Amico DJ, Baker AS.Endogenous bacterial endophthalmitis: report of a 10-yearretrospective study. Ophthalmology 1994;101(5):832–8.

82. Khan A, Okhravi N, Lightman S. The eye in systemic sepsis. Clin Med2002;2:444–8.

83. Lingappan A, Wykoff CC, Albini TA, et al. Endogenous fungalendophthalmitis: causative organisms, management strategies, andvisual acuity outcomes. Am J Ophthalmol 2012;153:162–6.

84. Feman SA, Nichols JC, Chung SM, Theobald TA. Endophthalmitis inpatients with disseminated fungal disease. Trans Am OphthalmolSoc 2002;100:67–72.

85. Dozier CC, Tarantola RM, Jiramongkolchai K, Donohue SP. Fungaleye disease at a tertiary care center: the utility of routine inpatientconsultation. Ophthalmology 2011;118:1671–6.

86. Shen X, Xu G. Vitrectomy for endogenous fungal endophthalmitis.Ocul Immunol Inflamm 2009;17:148–52.

87. Connell PP, O’Neill EC, Fabinyi D, et al. Endogenousendophthalmitis: a 10 year experience at a tertiary referral center.Eye (Lond) 2011;25:66–72.

88. Karmisholt MK, Hjort U, Knudsen LL, Schonheyder HC. Candidaemiaand risk of intraocular infection: a Danish hospital-based cohortstudy. Scand J Infect Dis 2008;40:241–6.

89. Oude Lashof AM, Rothova A, Sobel JD, et al. Ocular manifestationsof Candidemia. Clin Infect Dis 2011;53(3):262–8.

90. Shah CP, McKey J, Spirn MJ, Maguire J. Ocular candidiasis: areview. Br J Ophthalmol 2008;92:466–88.

91. Ruhnke M, Rickerts V, Cornely OA, et al. Diagnosis and therapy ofCandida infections: joint recommendations of the German SpeakingMycological Society and the Paul-Ehrlich-Society for Chemotherapy.Mycoses 54:279–310.

92. Connell PP, O’Neill EC, Amirul Islam FM, et al. Endogenousendophthalmitis associated with intravenous drug abuse. Seven-year experience at a tertiary referral center. Retina 2010;30:1721–5.

93. Rao NA, Hidayat AA. Endogenous mycotic endophthalmitis:variations in clinical and histopathologic changes in candidiasiscompared with aspergillosis. Am J Ophthalmol 2001;132:244–51.

94. Keyashian K, Malani PN. Endophthalmitis associated withintravenous drug use. S Med J 2007;100:1219–20.

95. Omuta J, Uchida K, Yamaguchi H, Shibuya K. Histopathologicalstudy on experimental endophthalmitis induced by bloodstreaminfection with Candida albicans. Jpn J Infect Dis 2007;60:33–9.

96. Essman TF, Flynn Jr HW, Smiddy WE, et al. Treatment outcomes ina 10-year study of endogenous fungal endophthalmitis. OphthalmicSurg Lasers 1997;28:185–94.

97. Willermain F, Bradstreet C, Kampouridis S, et al. Differentpresentations of ophthalmic aspergillosis. Eur J Ophthalmol 2008;18(5):827–30.

98. Riddell J, McNeil SA, Johnson TM, Bradley SF, Kazanjian PH,Kauffman CA. Endogenous Aspergillus endophthalmitis: report of 3cases and review of the literature. Medicine 2002;81:311–20.

99. Wong JS, Chan TK, Lee HM, Chee SP. Endogenous bacterialendophthalmitis: an east Asian experience and a reappraisal of asevere ocular affliction. Ophthalmology 2000;107:1483–91.

100. Chen KJ, Hwang YS, Chen YP, Lai CC, Chen TL, Wang NK.Endogenous Klebsiella endophthalmitis associated with Klebsiellapneumoniae Pneumonia. Ocul Immunol Inflamm 2009;17:153–9.

101. Chen YJ, Kuo HK, Wu PC, et al. A 10-year comparison ofendogenous endophthalmitis outcomes: an east Asian experiencewith Klebsiella pneumoniae infection. Retina 2004;24:383–90.

102. Christensen SR, Hansen AB, La Cour M, Fledelius HC. Bilateralendogenous bacterial endophthalmitis: a report of four cases. ActaOphthalmol Scand 2004;82:306–10.

103. Greenwald MJ, Wohl LG, Sell CH. Metastatic bacterialendophthalmitis: a contemporary reappraisal. Surv Ophthalmol1986;31(2):81–101.

104. Lee SY, Chee SP. Group B Streptococcus endogenousendophthalmitis: case reports and review of the literature.Ophthalmology 2002;109:1879–86.

105. Almeida DR, Miller D, Alfonso EC. Anterior chamber and vitreousconcordance in endophthalmitis. Implications for prophylaxis. ArchOphthalmol 2010;128:1136–9.

106. Sugita S, Kamoi K, Ogawa M, Watanabe K, Shimizu N, Mochizuki M.Detection of Candida and Aspergillus species DNA using broad-range real-time PCR for fungal endophthalmitis. Graefes Arch ClinExp Ophthalmol 2012;250:391–8.

107. Bispo PJ, de Melo GB, Hofling-Lima AL, Pignatari ACC. Detectionand Gram discrimination of bacterial pathogens from aqueous andvitreous humor using real-time PCR assays. Invest Ophthalmol VisSci 2011;52:873–81.

108. Biwas J, Bagyalakshmi R, Therese LK. Diagnosis of Aspergillusfumigatus endophthalmitis from formalin fixed paraffin-embeddedtissue by polymerase chain reaction-based restriction fragmentlength polymorphism. Indian J Ophthalmol 2008;56:65–6.

109. Horton JC, Yoon MK, Carvalho MD, McLeod SD. Polymerase chainreaction confirmed by immunohistochemistry: a two-prongedapproach in endophthalmitis. Acta Ophthalmol 2011;89:301–2.

110. Marak Jr GE. Phacoanaphylactic endophthalmitis. Erratum in: SurvOphthalmol 1992;36:454.

111. Hochman M, Sugino IK, Lesko C, Friedman AH, Zarbin MA.Diagnosis of phacoanaphylactic endophthalmitis by fine needleaspiration biopsy. Ophthalmic Surg Lasers 1999;30:152–4.

112. Thach AB, Marak Jr GE, McLean IW, Green WR. Phacoanaphylacticendophthalmitis: a clinicopathologic review. Int Ophthalmol1991;15:271–9.

113. Mamalis N, Edelhauser HF, Dawson DG, Chew J, LeBoyer RM,Werner L. Toxic anterior segment syndrome. J Cataract Refract Surg2006;32:324–33.

114. Holland SP, Morck DW, Lee TL. Update on toxic anterior segmentsyndrome. Curr Opin Ophthalmol 2007;18:4–8.

115. Patnaik B, Biswas C, Patnaik RK. Sterile endophthalmitis invitrectomised eyes due to suspected heat resistant endotoxins inthe infusion fluid. Indian J Ophthalmol 2004;52:127–31.

116. Stepien KE, Eaton AM, Jaffe GJ, Davis JL, Raja J, Feuer W.Increased incidence of sterile endophthalmitis after intravitrealtriamcinolone acetonide in Spring 2006. Retina 2009;29:207–13.