SPECIAL CORNEA ISSUE
Contact Lenses & Keratitis: What's the Connection?
Experts describe the process that leads to infection — and how to manage it clinically
BY JESSE PELLETIER, MD & JOSEPH CAPRIOTTI, MD
Recent publicity over the well-documented increase in Fusarium keratitis that occurred in 2005-2006 and the Acanthamoeba outbreak described in 2007 have led the CDC, the eye care community and the public to re-focus their attention on the relationship between contact lenses and corneal infections. Though the vast majority of the 130 million1 regular contact lens wearers will never experience any serious complications, occurrence in even a small percentage of these users represents a significant dilemma for ocular public health. There is no question that contact lens users have a higher incidence of corneal infections and this increased risk persists in all users of all lens types.2
We have learned a tremendous amount in the last three decades about the infection risks associated with soft contact lens use. By examining our collective experience, specifically with CLs and stromal keratitis, we hope to provide a broad answer to the question, "How does contact lens wear cause an increased incidence of infectious keratitis?" We discuss below some of the general mechanisms the eye has to defend against surface bacterial infections and how these defenses are compromised in the setting of soft contact lens use.
Lens Type and Risk of Infection
Lens-related complications in earlier generations of contact lenses were thought to be related to hypoxia. These first commercial CLs approved in the US employed relatively high Dk (a measure of O2 permeability) materials such as polymethyl- methacrylate that led to a variety of hypoxia-related problems. With the introduction of the first hydrogel lenses in 1971, and further with the launch of silicone hydrogels in 1999, it was indeed found that many undesirable CL effects were minimized. The bacterial keratitis rate, however, even in high-Dk silicone hydrogel lens users, remained unequivocally higher.2,3
Thus, despite the delivery of physiologic levels of oxygen through the relatively high-Dk lenses, bacterial keratitis remained an infrequent but serious complication.4 Accordingly, with hypoxia now effectively eliminated as the sole cause of the increased infection rate, other aspects of surface health were examined.
Though the complete picture remains elusive, there seem to be aspects of the natural corneal defense system that are compromised simply by the presence of a contact lens.
How Does the Cornea Normally Prevent Bacterial Infection?
The ocular surface is under constant assault by microbial, mechanical and photochemical elements. It is among the most highly exposed and critically maintained mucosal surfaces in the body — even a small scar in the central cornea is vision threatening. The transparency of the ocular surface requires the absence of vascularity, keratinized epithelium and the other principal defense mechanisms available to other exposed surfaces (i.e., the hands, facial skin, etc.). It is only when bacteria violate the integrity of the epithelium and gain access to the stroma that a clinically apparent bacterial keratitis develops. These stromal infections are what commonly present as ulcerative keratitis with robust inflammatory components, tissue destruction and stromal necrosis.
Obviously trauma is a frequent cause of bacterial translocation from the surface to the stroma, but most CL-related infections are not the result of apparent trauma, leading us to conclude that bacteria are somehow accessing the stroma through clinically indiscernable microtrauma or an intact epithelium. So what exactly prevents bacteria from normally accessing the stroma? And how are these defenses modified with contact lens use?
The Tear Film
The tear film, of course, forms the first external barrier and the primary biochemical means of surface defense. The precise anti-bacterial mechanisms of tear fluid remain elusive, but the involvement of mucin, IgA, bacteriostatic enzymes and anti-infective peptides all seem to be important.5 Tear fluid alone can prevent certain Pseudomonas aeruginosa strains from invading or killing cultured corneal cells in vitro.6 In addition, the variety of chemical and photoactive factors present, there appears to be a mechanical effect that fully-constituted tears have in maintaining a healthy infection-free surface.
Working with the lids and the blink reflex, the tear film is responsible for the periodic flushing of the ocular surface, sweeping away chemical, microbial and cellular debris. It is well known that conditions that affect either the production (e.g., chronic dry eye, Sarcoidosis, lacrimal pathology) or distribution (e.g., ectropion, exophthalmos, lid retraction) of the tear film carry an increased risk of corneal infection.
The Corneal Epithelium
The multi-layered, non-keratinized, avascular, transparent epithelium of the corneal surface lies just beneath the tear film. It is characterized by distinct permeability differences on its anterior (apical) and posterior (basal) aspects and a discrete network of tight junctions that effectively isolate both poles.
While the anterior surface is much more resistant to bacterial penetration, the basal surface is easily compromised, particularly by the common CL-related strains of Pseudomonas. It is to the eye's advantage, then, to prevent the commonly colonized apical surface from contacting the easily violated basal surface. This apical/basal isolation accomplished by intact tight epithelial junctions forms a highly protective antibacterial arch itecture. Tight junctions further enable selective removal of individually infected cells through sloughing (also enhanced by the actions of the tear film/lid/blink system).
The epithelial barrier function exists in addition to a complex of immune activity. As a constituent of the innate immune system, the epithelium can detect, identify and modulate responses to bacterial pathogens, largely through the toll-like receptor system.7
It is important to note that epithelial injury is not necessarily a prelude to infection. In fact, in many cases of chronic epithelial breakdown (e.g., keratoconus, recurrent erosion syndromes, some corneal dystrophies, HSV epithelial keratitis) there is no (or a very small) increased risk of stromal keratitis. This, along with similar results in controlled animal experiments of epithelial injury and inoculation,8 suggest that the epithelium must have some other protective mechanism that is redundant with the protection afforded by an intact surface.
Basal Lamina
Basal epithelial cells secrete extracellular matrix material between the epithelium and Bowman's layer in the human cornea. This layer forms a nanometer-scale three-dimensional reticular structure well characterized by atomic force microscopy.9 The pores in this reticular network are much smaller than invasive bacteria and form an additional mechanical defense against intracellular invasion.
Aside from the obvious barrier effects conferred by the basal epithelial layer, small variations in cell surface topology have a role — independent of biochemical mediators — in the recruitment of immune activators. The basal lamina provides another two-pronged defense employing both structural and biochemical elements. Animal experiments have shown that epithelial damage alone, with intact basal lamina, is an ineffective means for induction of stromal keratitis, even with the application of pathogenic organisms to the defect.
A Practical Approach to Managing CL KeratitisBY JESSE PELLETIER, MDMy treatment of infectious contact lens keratitis begins with a thorough history and review of risk factors. I then determine whether the infiltrate is vision threatening and/or located in the central corneal zone. Peripheral corneal lesions are more likely to be non-infectious, and the increased corneal thickness makes rapid perforation rare. Next, characterization of the infiltrate is important, as it may shed light on the nature of the offending microbe. I look for depth of ulceration, keratolysis with or without suppuration, density/size/number of infiltrate(s), shape, feathery edges, immune rings, perineural infiltrates, dendritic lesions and other tell-tale signs. Also, evaluation for associated hypopyon and scleritis is crucial, as these can be indicators of a more serious infection that may require additional treatments such as oral antibiotics and NSAIDs. Fortunately, most cases of bacterial keratitis will resolve with empiric, broad-spectrum therapy and with the patient retaining good visual function. For those infiltrates that are sight threatening, chronic, atypical in nature or unresponsive to monotherapy, cultures and corneal scrapings remain invaluable tools for microbe identification. Moreover, one can debulk a shallow corneal infiltrate and decrease bacterial load utilizing a blade or forceps. I routinely culture on blood, chocolate, Sabouraud's media, agar-agar, thioglycollate media and perform corneal smears for direct microbe identification. If the patient continues to worsen, and an organism is not identifiable, re-culture is appropriate and occasionally corneal biopsy is as well. In my hands, treatment of severe bacterial keratitis (i.e., dense, central ulceration with suppuration) involves fortified antibiotics such as vancomycin and tobramycin with frequent (e.g., hourly) dosing for effective corneal saturation. Other commonly employed strategies include broad-spectrum 8-methoxy fluoroquinolone monotherapy, high concentration fluoroquinolone monotherapy or a fourth-generation quinolone in conjunction with a fortified antibiotic. Once the therapeutic regimen is underway, clinical monitoring is of the essence to gauge the host response. As the patient improves and depending on culture results, I consider switching to monotherapy, tapering drops while maintaining therapeutic levels, and utilizing adjunctive agents. I think it is important to add steroids early in culture-positive cases (excluding mycobacteria), usually 24 to 48 hours after the patient has begun topical antibiotic therapy. The goal here is to reduce immune-related sequelae and long-term visual loss. Atypical microbial pathogens such as Acanthamoeba can be among the most difficult infections to diagnose and treat. Early detection remains crucial in the setting of any atypical keratitis partly because our armamentarium is not sufficiently advanced to reproduce the good outcomes associated with bacterial infections. My suspicions for AK are heightened in the face of any of the well-known risk factors, or confirmation of any of the early or late corneal findings previously described. Diagnosis is usually made with corneal culture on agar-agar with non-nutrient E. coli overlay. Here one can directly observe the trophozoites under direct microscopy migrating from the agar scarification site. Other diagnostic methods may include a microbe staining, confocal microscopy, PCR and corneal biopsy. In some cases, however, it is necessary to treat patients with a provisional diagnosis of AK based on suspicious biomicroscopic findings alone and/or known risk factors. Current medical treatment revolves around the utilization of two classes of medicines, the biguanides and diamides. Because of difficulty in obtaining the latter, I often start my patients on monotherapy PHMB 0.02% to 0.06% dosed hourly. Chlorhexidine is another option, however, I find it to be more epithelial toxic than PHMB. Finally, Brolene (a diamide) is difficult to obtain but used in conjunction with a biguanide for initial therapy by many corneal specialists outside the US. There are other medicines that have been used in AK such as neomycin, and various anti-fungals, however their activity against acanthamoeba cysts remains dubious. The role of steroids in AK remains controversial and I do not use them in disease confined to the cornea. However, oral steroids in cases of AK associated scleritis may be of benefit. Recalcitrant cases of AK should be examined for co-infection with other microbes including HSV and possible resistance to the cysticidal agent of choice should be considered. |
Contact Lens Effects on Surface Defense Mechanisms
We know from a variety of studies in animal models that the most effective way to cause a bacterial keratitis is to inoculate the stroma, and we have just described three effective components of the healthy ocular surface that provide an overlapping system of physical, mechanical and chemical protection for the eye. We turn now to consider how CL use may disrupt the natural surface defenses and provide bacteria an avenue to the stroma.
The tear film depends on synthetic, excretory and regulatory activity from disparate sites around the eye. The aqueous component, produced in the lacrimal system, must be effectively mixed with mucinous elements produced by conjunctival goblet cells, lipid secretions from the lid margin Meibomian apparatus and chemical species produced in epithelial sites and recruited from peripheral conjunctival vasculature. This all has to be evenly spread by the lids, modulated by the blink reflex and turned over through the drainage cycle. A closely applied synthetic material that compromises any of these functions is likely to compromise tear film defense. Contact lenses, including silicone hydrogel soft lenses, easily disrupt this system.
Soft lenses in particular trap the tear film and inhibit tear film recycling. The balanced three-layer structure can be separated into lipid/mucin/aqueous components with clusters of the mucin component under the lens ("mucin balls"). Prolonged CL wear causes desensitization of corneal nerves and can impair the corneal blink reflex. Chemical mediators can be diluted and even absorbed into hydrophilic lenses, diluting their concentration and decreasing their efficacy. The sloughing function of the lid/blink/tear film system is impaired and infected epithelial cells are more difficult to eliminate.
Epithelial damage from contact lenses is easily seen via fluorescein staining. The contact lens is a chronic foreign body in constant motion across the epithelial surface. Although frank abrasions are rare, minor epithelial trauma is nearly universal in CL wearers and easily identified at the slit lamp. In addition to the lens itself, cellular, chemical and microbial debris is trapped between the lens and the epithelial surface in the precise space where tear film exchange is restricted by smooth, soft lenses.
Once the tear film is compromised, surface homeostasis can become dysregulated with derangement effects on the underlying epithelium and basal lamina. The precise microenvironment of each individual epithelial cell is altered with likely extracellular structural changes and intracellular metabolic sequelae. The production and modulation of biochemical mediators is certainly altered with unknown consequences for epithelial surface chemistry and basal lamina structure and function. The domino effect of poor tear film exchange ends with disruption of the remaining defenses in the epithelium and basal lamina. The final result is a bacterial highway to the stroma and a boulevard of broken dreams for the stromal keratitis patient.
Common Bacterial Pathogens
By a significant margin, Pseudomonas aeruginosa has received the most attention as a cause of infectious keratitis in CL wearers. This is certainly warranted, as contact lens wear is a major risk factor for this infection, which may cause severe, fulminating keratitis.10,11 Pseudomonas species have a variety of virulence factors at their disposal to inflict maximum damage. Most commonly noted are specialized proteases, biofilm formation and fluoroquinolone resistance, which all play a role in highly cytotoxic strains.12 There is even evidence that certain strains can damage epithelia on an uninjured corneal surface given prolonged exposure time.13
Virulent contact-lens related keratitis requires aggressive broadspectrum therapy to limit risks of long-term visual impairment.
Given the amount of attention lavished on Pseudomonas, it is important to remember that there are other bacterial organisms commonly implicated, but less frequently discussed, that can cause contact lens-related keratitis. These include Staphylococcus, Streptococcus and Serratia species. In fact, a recent study on contact lens-induced keratitis in Japan identified gram positive bacteria as being the most commonly isolated bacteria (82.9%).14 This reinforces that vigilance, broad suspicion and the proper methods of identification should be employed in the more serious cases of CL keratitis.
Atypical Pathogens in Contact Lens Keratitis
The fungal keratitis outbreak beginning in 2005 was not confined to the United States. Other countries, such as Hong Kong and Singapore, published early studies and helped alert the US to concerns regarding the ReNu with MoistureLoc multi-purpose contact lens solution. This chain response ultimately lead to its withdrawal from the global market in 2006.15,16 Etiologies responsible for this outbreak were thought to be multi-factorial and ranged from poor hygiene to sequelae secondary to the Asian tsunami and hurricane Katrina, with eventual thermal-related solution instability being the most recently evaluated postulate.17
Fungal keratitis most commonly arises from molds such as Fusarium and Aspergillus, or yeasts such as Candida. During the outbreak, Fusarium species were the reported perpetrators. In general, fungal infection requires a breach in corneal epithelial integrity, and an adequate organism inoculate. Once an established stromal infection has occurred, it may take the commonly described form of hyphal extension (feathery edges) and satellite lesions. Classically difficult to eradicate and late to diagnose, fungal keratitis often requires surgical intervention to halt disease, as spread into the anterior chamber and/or sclera may cause irreversible ocular morbidity. Traditional treatment options include the polyenes, such as natamycin and amphotericin B, which characteristically demonstrate poor corneal penetration and efficacy. Newer triazole antifungal agents such as voriconazole and posaconazole may offer greater promise and their use in these infections is a continued area of interest.
Acanthamoeba keratitis (AK) with contact lens use dramatically increased in the mid-1980s and this trend has been reprised during the past few years in both the US and abroad. Acanthamoeba are protozoans that live freely in the soil, air, dust and water. A. castellani and A. polyphaga are the most common of the eight species implicated in keratitis.18 Risk factors associated with infection include poor hygiene, swimming in CLs, poor disinfection, tap water exposure, susceptible contact lens solutions, overnight orthokeratology and minor corneal trauma.19
A major obstacle in treating AK lies in the diagnostic delay that usually accompanies these cases, allowing for greater ocular penetration of the amoeba. Moreover, the cystic element of the organism is resistant to most chemicals while eradication requires toxic therapy with multiple-drug regimens. In any case, despite prolonged therapy after definitive diagnosis, there is still an increased rate of progression to surgery when compared with bacterial keratitis.
AK infections may take multiple forms; classic descriptions include pain out of proportion to examination, radial perineuritis and ring ulceration. Among corneal specialists who treat these infections with some frequency, it is thought that early recognition is key. Early AK manifestations commonly overlooked include punctate keratopathy, pseudodendrites, subepithelial and perineural infiltrates. In fact, if the infection is recognized and treated while confined to epithelial disease, good visual outcomes are within reason.18,19
Conclusion
The ocular surface is designed with redundant defenses to protect against infectious assault from a variety of native and external threats. When these defenses are compromised, exposure to normally tolerable insults can rapidly induce pathogenic events. There is no doubt that contact lens use carries an increased risk of corneal infection, likely from disrupting the balance between microbe and host defense.
Despite advances in lens materials, ophthalmic antibiotic therapy and diagnosis, bacterial and atypical keratitides are important considerations in the ongoing care of your lens-wearing patients. Attention to subtle clinical signs and frank discussion with contact lens users can be an effective way to build a valuable partnership to prevent, recognize and respond to lens-related infections. OM
References
1. Barr, J. "2004 Annual Report". Contact Lens Spectrum. January, 2004.
2. Dart JKG, Radford CF, Minassian D, Verma S, Stapleton F. Risk Factors for Microbial Keratitis with Contemporary Contact Lenses: A Case-Control Study, Ophthalmology. 2008;115:1647-1654.
3. Schein OD, McNally JJ, Katz J, Chalmers RL, Tielsch JM, Alfonso E, Bullimore M, O'Day D, Sholvin J. The incidence of microbial keratitis among wearers of a 30-day silicone hydrogel extended-wear contact lens. Ophthalmology. 2005;112:2172-2179.
4. Stapleton F, Stretton S, Papas E, Skotnisky C., Sweeney DF. Silicone hydrogel contact lenses and the ocular surface. Ocul Surf. 2006;4:24-43.
5. McNamara NA, Andika R, Kwong M, Sack RA, Fleiszig SM. Interaction of Pseudomonas aeruginosa with human tear film components. Curr Eye Res. 2005;30:517-525. Invest Ophthalmol Vis Sci. 2007;48:4664–4670.
6. Fleiszig SM, Zaidi TS, Pier GB. Mucus and Pseudomonas interactions with human tear fluid. Infect Immun. 2003;71:3866-3874.
7. Kumar A, Yin J, Zhang Z, Yu F. Modulation of Corneal Epithelial Innate Immune Response to Pseudomonas Infection by Flagellin Pretreatment. Invest Ophthalmol Vis Sci. 2007;48:4464-4470.
8. Lee EJ, Evans DJ, Fleiszig SM. Role of P. Aeruginosa ExsA in penetration through corneal epithelium in a novel in vivo model. Invest Ophthalmol Vis Sci. 2003;4:5220-5227.
9. Abrams GA, Schaus SS, Goodman SL, Nealey PF, Murphy CJ. Nanoscale topography of the corneal epithelial basement membrane and Descemet's membrane of the human. Cornea. 2000;19:57-64.
10. Dart JK. Predisposing factors in microbial keratitis; the significance of contact lens wear. Br J Ophthalmol. 1988;72;926-930.
11. Keay L, Edwards K, Naduvilath T, et. al. Microbial keratitis; predisposing Factors and Morbidity. Ophthalmology. 2006;113:109-116.
12. Choy M, Stapleton F, Willcox M, et. al. Comparison of virulence factors in Pseudomonas aeruginosa strains isolated from contact lens- and non-contact lens-related keratitis. J Med Microbiol. 2008;57:1539-1546.
13. Fleiszig SM, Lee EJ, Wu C. Cytotoxic strains of Pseudomonas aeruginosa can damage the intact corneal surface in vitro. CLAO J. 1998;24;41-47.
14. Inoue N, Toshida H, Mamada N, et. al. Contact lens-induced infectious keratitis in Japan. Eye Contact Lens. 2007;33:65-69.
15. Ma SK, So K, Chung PH, et. al. A multi-country outbreak of fungal keratitis associated with a brand of contact lens solution: the Hong Kong experience. Int J Infect Dis. 2009;13;443-448.
16. Khor WB, Aung T, Saw SM, et. al. An outbreak of Fusarium keratitis associated with contact lens wear in Singapore. JAMA. 2006;295:2867-2873.
17. Bullock JD. Root cause analysis of the Fusarium keratitis epidemic 2004-06 and prescriptions for preventing future epidemics. Trans Am Ophthalmol Soc. 2009;107:194-204.
18. Dart JK, Saw VP, Kilvington S. Acanthamoeba keratitis: diagnosis and treatment update 2009. Am J Ophthalmol. 2009;148:487-499.
19. Por YM, Mehta JS, Chua JL. Acanthamoeba keratitis associated with contact lens wear in Singapore. Am J Ophthalmol. 2009;148:7-12.
| Jesse Pelletier, MD is a founding partner and director of cornea, cataract and refractive surgery at the Ocean Ophthalmology Group in Miami, voluntary assistant professor of ophthalmology at Bascom Palmer Eye Institute, and an attending ophthalmologist at the Miami VA. |
|
Anthony Capriotti, MD is an ophthalmologist and research scientist with an interest in ocular microbiology, infection and inflammation. He is associate research director of the Ocean Ophthalmology Group and an adjunct scientist in the department of chemistry at Columbia University. |
![]({"src":"/media/dniiolyx/62010.jpg","focalPoint":null,"crops":[{"alias":"thumbnail","width":1110,"height":700,"coordinates":null}]})