Microbial Keratitis

Microbial Keratitis and Contact Lens Wear

This review examines past and current findings on the condition and how it affects contact lens wearers.

By Katherine Weibel, OD; William L. Miller, OD, PhD, FAAO; & Jason J. Nichols, OD, MPH, PhD, FAAO

Microbial keratitis (MK) is an ocular emergency (Thomas and Geraldine, 2007), characterized by a corneal epithelial defect with an underlying stromal infiltrate (Keay et al, 2006; Pachigolla et al, 2007; Willcox, 2012). Immediate and intense antimicrobial treatment is necessary to prevent vision loss (Keay et al, 2006; Pachigolla et al, 2007; Thomas and Geraldine, 2007; Willcox, 2012); however, even with treatment, MK can result in corneal perforation, scarring, endophthalmitis, and even eye loss (Pachigolla et al, 2007).

MK is caused by replicating organisms such as bacteria, viruses, fungi, and parasites (Keay et al, 2006). It is critical to determine the causative agent so that the proper therapy may be instituted immediately. Keratitis may result from numerous causes including viruses such as herpes simplex, auto-immune disease, and vernal keratoconjunctivitis (Srinivasan et al, 2008).

In this article, we will focus on the types of keratitis that are most associated with contact lens wear and how to treat them.

Bacterial Keratitis

Despite the abundance of contact lenses that offer high oxygen transmission and the common use of frequent replacement lens modalities, bacterial corneal infections related to contact lens wear persist. Bacteria are the most common organism associated with MK in developed countries (Alexandrakis et al, 2000; Keay et al, 2006). Additionally, contact lens wear is among the greatest risk factors for bacterial keratitis in developed countries (Kaye et al, 2010).

Studies from the United Kingdom and the United States report that the most common gram-positive pathogens associated with bacterial keratitis include coagulase negative staphylococci and Staphylococcus aureus (Alexandrakis et al, 2000; Kaye et al, 2010; Sueke et al, 2010). Gram-positive bacteria usually produce small, discrete, abscess-like lesions. Ulcers caused by Staphylococcus may produce less pain, grow slowly, and have a quieter presentation when compared to gram-negative infections (Srinivasan et al, 2008). Likewise in the United Kingdom and the United States, the most commonly reported gram-negative bacteria associated with bacterial keratitis are Pseudomonas aeruginosa and the Enterobacteriaceae family, which includes Serratia marcescens (Alexandrakis et al, 2000; Kaye et al, 2010; Sueke et al, 2010).

In contrast to gram-positive infections, gram-negative bacteria, such as Pseudomonas, are more likely to cause diffuse and quickly spreading lesions (Srinivasan et al, 2008). Further, gram-negative infections are associated with large infiltrates and severe anterior chamber reactions (Bourcier et al, 2003). Additionally, tearing, pain, and vision loss are typically more severe in Pseudomonas-caused infections. Significant eyelid and conjunctival edema and pus-filled discharge are more often associated with gram-negative infections (Srinivasan et al, 2008).

Bacterial keratitis requires immediate and aggressive therapy, often in the form of monotherapy with a topical fluoroquinolone or a fortified antibiotic combination (Constantinou et al, 2007). A limitation of a fluoroquinolone may include increasing bacterial resistance, while fortified antibiotics may be less readily available, more costly, and possibly more toxic to the ocular surface (Constantinou et al, 2007).

A recent study comparing moxifloxacin (1.0%), ofloxacin (0.3%), and fortified tobramycin (1.33%)/cephazolin (5.0%) found all three antibiotics to be equally effective against bacterial organisms, with no significant differences in the average time to cure or with complications such as perforation or enucleation. Furthermore, over the study period there were no significant differences among the three groups in the severity of biomicroscopy findings such as conjunctival discharge, bulbar conjunctival injection, superficial punctate keratopathy, infiltrate intensity, or anterior chamber reaction (Constantinou et al, 2007). Similarly, a study comparing gatifloxacin (0.3%), moxifloxacin (0.5%), and tobramycin 1.3%/cephazolin (5.0%) found no significant difference in the three groups in treatment efficacy, ulcer healing time, final visual acuity, or serious complication incidence such as perforation or non-healed ulcers (Shah et al, 2010). It’s important to note that both of these studies were conducted outside of the United States and that outcomes may differ in the United States as common bacteria vary in virulence by country.

In contrast, the use of ciprofloxacin or ofloxacin as monotherapy treatment may not be appropriate for patients presenting with large corneal ulcers in which Streptococcus species is either suspected or identified with a culture (Kaye et al, 2010). A laboratory study from the United States provides more information on fluoroquinolones and the susceptibility of bacteria. This study found that the four fluoroquinolones tested demonstrated consistent activity against methicillin-susceptible Staphylococcus aureus, Streptococcus pneumoniae, and Haemophilus influenzae bacteria isolated from ocular infections (Asbell et al, 2008). Susceptibility ranged from approximately 80 percent to 100 percent, depending on the bacteria and fluoroquinolone tested. In contrast, most methicillin-resistant Staphylococcus aureus (MRSA) bacteria displayed a high degree of in vitro resistance to the fluoroquinolones, with a susceptibility of only 15.2 percent. The authors concluded that if MRSA is a likely causative organism, treatment other than fluoroquinolones should be considered (Asbell et al, 2008). Before a treatment is chosen, practitioners must carefully consider the eye’s clinical appearance, the infiltrate’s size and location, and other concurrent risk factors. Further studies are necessary to determine whether new fluoroquinolones are as effective compared to standard fortified antibiotics in treating bacterial keratitis (Wong et al, 2012).

Fungal Keratitis

Although this type of corneal infection is more prevalent in tropical areas and with younger males involved in agriculture (Srinivasan et al, 2008), contact lens wearers may develop the historically rare fungal infection (Alfonso et al, 2006; Srinivasan et al, 2008). Fungal keratitis may also affect those who have chronic ocular surface disease, immunocompromised health status, diabetes, and those who use steroids (Srinivasan et al, 2008).

Although typically rare for cosmetic contact lens wearers, fungal infections can cause destructive outcomes for vision. Prompt and accurate diagnosis and treatment of fungal keratitis has been associated with faster resolution times (Alfonso et al, 2006). Fungal keratitis may be characterized by a stromal infiltrate, hypopyon, pain, and decreased vision (Alfonso et al, 2006). The infiltrate may appear raised, wet, and grayish-white or yellowish-white. Satellite lesions may exist (Srinivasan et al, 2008).

A well-known outbreak of Fusarium keratitis was reported in 2005 and 2006 in Asia and also in the United States (Khor et al, 2006). A study from Singapore found common characteristics of the Fusarium keratitis outbreak with disposable soft lens wear (98.5 percent), deficient contact lens hygiene (81.8 percent), and the prior use of Renu (Bausch + Lomb) multipurpose lens care solutions (93.9 percent). Of the Renu users, 63.6 percent reported using Renu with MoistureLoc (Khor et al, 2006). In 2006, B+L conducted a worldwide recall of Renu with MoistureLoc, and Fusarium keratitis has declined significantly (Tu and Joslin, 2010).

Treatments for fungal keratitis may include natamycin, fluconazole, amphotericin B, and miconazole (Al-Badriyeh et al, 2010). Topical natamycin is widely used and the only commercially available medication for fungal keratitis. However, it has poor penetration into the deeper cornea and may be a suitable treatment for superficial fungal keratitis in the early stages, but likely not for advanced fungal keratitis affecting the deeper layers of the cornea (Pradhan et al, 2011). Because the traditional treatments mentioned have shortcomings such as poor corneal penetration, limited effectiveness against various fungi, and limited clinical usefulness, newer antifungals, such as voriconazole, are being investigated (Al-Badriyeh et al, 2010) and currently compounded for ocular use.

Amoebic Keratitis

Acanthamoeba is an organism present in air, soil, dust, and water. This protozoa has a life cycle that includes trophozoite (mobile and active) and cyst stages (Dart et al, 2009). The cystic stage allows Acanthamoeba to survive in adverse environments (Dart et al, 2009). Acanthamoeba keratitis (AK) occurs annually in about one to two cases per million contact lens wearers (Schaumberg et al, 1998). Approximately 85 percent of Acanthamoeba keratitis is associated with contact lens wear in countries where lens wear is common (Dart et al, 2009). Acanthamoeba keratitis is potentially sight-threatening and has strong associations with soft contact lens wear and improper contact lens hygiene, such as topping off old lens care solution (Verani et al, 2009). The risk of Acanthamoeba keratitis may increase with swimming while wearing contact lenses, close proximity to water sources that are stagnant or experiencing flooding, and in warmer months (Ibrahim et al, 2007).

Verani et al (2009) reported that beginning in 2004, a gradual rise in AK occurred in the United States. This study on the outbreak of AK in the United States found that 89 percent of those patients who developed AK were contact lens wearers, of whom 88 percent wore soft lenses. Furthermore, the use of the multipurpose lens care solution Complete Moisture-Plus (Advanced Medical Optics, now Abbott Medical Optics) was associated with approximately a 16 times risk factor for developing AK. The disinfection solution was hypothesized to be inadequate against Acanthamoeba and was recalled in 2007 (Verani et al, 2009).

While neither the Fusarium nor the Acanthamoeba outbreaks are completely understood, the inadvisable practice of topping off contact lens care solutions was a substantial risk factor in both outbreaks (Verani et al, 2009).

Common characteristics of AK include pain, tearing, and photophobia, either unilaterally or sometimes bilaterally in contact lens wearers (Dart et al, 2009; Srinivasan et al, 2008). AK may start with punctate keratopathy, pseudodendrites, and perineural infiltrates. Perineural infiltrates develop along corneal nerves and may be evident during a detailed slit lamp examination. These infiltrates have a high association with AK (Dart et al, 2009). AK may progress in later stages to a ring infiltrate and anterior uveitis with hypopyon (Dart et al, 2009). A dense stromal ring infiltrate that spares the pupil area is often a diagnostic sign of AK (Srinivasan et al, 2008). Confocal microscopy may aid in the initial evaluation of AK; however, a definitive diagnosis can be made only with a culture or with polymerase chain reaction testing (Dart et al, 2009). Importantly, while the trophozoite phase of Acanthamoeba is susceptible to most antibiotics, anti-fungals, and antiprotozoals, the cystic phase of Acanthamoeba is usually not (Dart et al, 2009). Therefore, current thoughts on treatment include topical biguanides, such as polyhexamethylene biguanide (PHMB), and chlorhexidine, with or without the addition of diamidines (Dart et al, 2009).

While AK is usually rare, the treatment is costly. An Australian study in a teaching and research hospital setting found the cost of AK treatment to be the most expensive when compared to treatment for keratitis caused by gram-positive, gram-negative, and fungal pathogens (Keay et al, 2006). Notably, the cost to treat AK was more than triple the treatment cost for gram-negative bacteria-caused keratitis.

Even with medications, AK treatment can be a long and difficult process. In some reported cases, AK has shown no improvement even after 30 days of anti-Acanthamoeba eye medication use. Furthermore, AK medications such as PHMB and chlorhexidine typically are not readily available at most pharmacies and are usually compounded spontaneously (Khan et al, 2011). These eye medications are often made from industrial-quality sources, such as from swimming pool disinfectants, and may contain toxic impurities (Cavanagh, 2007).

Incidence of MK

The incidence of MK varies by study. Historically, for the United States, MK has been estimated at about 4.1 per 10,000 daily wear soft lens cosmetic wearers and 20.9 per 10,000 extended (overnight) soft lens cosmetic wearers (Poggio et al, 1989). Similarly, in the Netherlands, the incidence of MK was reported at 1.1 per 10,000 daily wear GP lens wearers, 3.5 per 10,000 daily wear soft lens wearers, and 20.0 per 10,000 extended wear soft lens wearers (Cheng et al, 1999). The advent of new lens materials, wear modalities, and an increasing use of contact lenses may have changed the arena of MK incidence. More recently in Australia, the annual incidence of MK among daily wear GP lens wearers was 1.2 per 10, 000 wearers, and for daily wear-only soft hydrogel wearers was 1.9 per 10,000 lens wearers. For daily disposable soft lens wearers (daily wear only), MK was estimated at 2.0 per 10,000 wearers, while for daily wear-only silicone hydrogel (SiHy) wearers, incidence was 11.9 per 10,000 wearers (Stapleton et al, 2008). Meanwhile, for wearers of extended wear hydrogel soft lens materials, conventional hydrogel materials, and planned replacement hydrogel materials, incidence was 19.5 while extended wear of silicone hydrogel material was 25.4, both per 10,000 lens wearers. In addition, a permanent decrease in best-corrected vision of ≥2 lines was reported in 13.9 percent of wearers who developed MK (Stapleton et al, 2008). In this same study, some of the risk factors associated with MK included extended (overnight) wear of lenses, poor lens case hygiene, smoking, and internet purchasing of lenses.

What is New?

Diagnosis While the slit lamp is a most helpful instrument for examining the eye, newer technology can examine ocular structures on a cellular level. The hope is for earlier and more accurate diagnosis of ocular conditions as they exist in life (in vivo). Early in vivo images of the human cornea using confocal microscopy were published in 1990 (Cavanagh et al, 1990). Because the condenser light source and the objective lens focus on the same precise point, the term “confocal” is used (Mantopoulos et al, 2010). The newest type of confocal microscopy, laser scanning confocal microscopy, utilizes a class 1 laser (Mantopoulos et al, 2010). Resolution varies among different confocal microscopes, but newer laser scanning confocal microscopes have reported 1μm/pixel capability (Niederer and McGhee, 2010). In vivo confocal microscopy allows the ocular surface to be assessed on a cellular level. Highly detailed images of the cornea, tear film, conjunctiva, and lids are possible (Zhivov et al, 2006).

Confocal microscopy has a vast number of clinical applications including, and not limited to, the imaging of corneal changes associated with contact lens wear, dry eye, keratoconus, corneal dystrophies, and MK (Niederer and McGhee, 2010). In vivo confocal microscopy is also helpful in diagnosing certain types of MK. In particular, the reliable and quick identification of Acanthamoeba is believed to be one of its most useful clinical applications. Both the trophozoite and cyst structures may be readily identified, aiding in rapid diagnosis and treatment of Acanthamoeba keratitis (Zhivov et al, 2006). Finally, confocal microscopy allows for the high resolution of ocular structures as they exist in a patient. This technology may serve as a beneficial connection between clinical and laboratory findings (Niederer and McGhee, 2010).

Treatment Due to concerning rates of microbial resistance and possible toxic side effects of available MK treatments, new avenues of treatment are under investigation (Cavanagh, 2007; Garg et al, 1999).

Corneal cross-linking is a technique that uses topical riboflavin (vitamin B2) and ultraviolet-A (UVA) irradiation to fortify the cornea and increase its biomechanical rigidity (Wollensak et al, 2003). In this process, riboflavin acts as a photosensitizer and is excited by UVA light. This interaction creates free radicals that result in the physical cross-linking and strengthening of collagen fibrils in the corneal matrix (Spoerl et al, 2004). Additionally important to the procedure, the UVA alone may have an effect against a variety of bacteria common to keratitis infections (Martins et al, 2008). Typically, the cross-linking procedure includes several steps. Topical anesthesia is used, and the epithelium of the infected area of the cornea is removed. Topical riboflavin 0.1% is instilled over the cornea approximately every three minutes for 30 minutes. The cornea is then irradiated with a UV light source of 370nm for 30 minutes (Garduno-Vieyra et al, 2011). Antibiotics may be prescribed or continued after the cross-linking procedure (Makdoumi et al, 2010; Spoerl et al, 2007) as can other antimicrobial medications and even a second cross-linking procedure if needed (Khan et al, 2011).

Corneal cross-linking procedures have demonstrated clinical success in MK treatment. In difficult-to-treat MK cases, cross-linking has provided dramatic improvement in patient symptoms, corneal epithelialization, clear anterior chamber, and significantly improved vision (Garduno-Vieyra et al, 2011; Khan et al, 2011; Makdoumi et al, 2010). The practitioners involved in these reports concluded that cross-linking may be an effective means of treating challenging MK cases, especially when confronted with infections not responding to usual treatment. However, further studies regarding corneal cross-linking are necessary (Khan et al, 2011; Makdoumi et al, 2010). It is important to understand that resolution with cross-linking typically takes several weeks. Finally, because corneal cross-linking uses UV light, unintentional eye damage is a valid concern. However, research regarding the safety of cross-linking shows that when specific guidelines are carefully followed, damage to the corneal endothelium, the lens, and the retina are not expected (Spoerl et al, 2007). Some of these guidelines include the correct application and concentration of topical riboflavin, use of a homogenous UV wavelength of 370nm, and a minimum corneal thickness of 400μm in the treated eye (Spoerl et al, 2007).

Conclusion

Microbial keratitis, a potentially devastating ocular emergency, continues to affect patients and contact lens wearers in particular. Ongoing research to improve diagnosis and treatment procedures provides an improved outlook for patients and practitioners alike. Education regarding correct contact lens care and wear schedules remains paramount for both wearers and prescribing eyecare practitioners. While immediate and accurate treatment remains the best defense against this serious condition, even better is its prevention. CLS

For references, please visit www.clspectrum.com/references.asp and click on document #211.