This article is partially an update of an article that I authored in the April 2011 issue of Contact Lens Spectrum titled “Microbiology and Contact Lens Wear.” As such, it will review what has been occurring in this area over the past six years. First, the good news: there have been no new epidemic outbreaks of contact-lens related microbial keratitis such as those related to a few multipurpose disinfecting solutions that occurred in 2006 and 2007.

However, as is inevitable in articles that discuss microbiology and contact lenses, I’m going to talk about the bad stuff: the rates of keratitis associated with contact lens wear and the types of causative microbes that are present on the ocular surface. I’ll review the major predisposing factors that can cause microbial keratitis (i.e., frank infection of the cornea during lens wear), the percentage of these cases that have been associated with contact lens wear from different countries, the risk factors associated with both microbial keratitis and corneal infiltrative events (i.e., those events that are associated with microbial contamination of lenses but do not result in frank infection of the cornea), the types and frequency of the microbes that have caused these, the antibiotic resistance of the microbes that cause microbial keratitis, and finally, I’ll include a discussion on the normal ocular microbiota.

MAJOR PREDISPOSING FACTORS FOR MICROBIAL KERATITIS AND THE MICROBES ISOLATED FROM THE CORNEA

The major predisposing factors for microbial keratitis are corneal trauma, systemic or ocular disease, and contact lenses (Tewari et al, 2012: Katara et al, 2013; Rahimi et al, 2015; Lin et al, 2015; and others. Full list available at www.clspectrum.com/references ). These reports have emanated from cases in the United States, Germany, Hong Kong, Taiwan, India, and Iran. Given the geographical and socioeconomic differences in these countries, it is perhaps not surprising that the percentage of cases associated with contact lens wear ranges from 0% to 5% in India and Iran (Tewari et al, 2012; Rahimi et al, 2015) to 31% in Taiwan (Lin et al, 2015) to 40% to 44% in the United States (Yildiz et al, 2012; Truong et al, 2016; Ni et al, 2015) and 43% to 83% in Hong Kong (Young et al, 2013; Ng et al, 2015).

Again, given the geographical and socioeconomic differences in these countries and the different predisposing factors for microbial keratitis, there are some differences in the spectrum of microbes that are isolated from the cornea during infection—the causative agents—although, in general, the types of bacteria associated with microbial keratitis are relatively consistent.

Of the cases reported from the United States, 32% to 69% of cases with microbial growth were caused by gram-positive bacteria, of which the most commonly isolated (10% to 53%, Table 1) were coagulase-negative staphylococci (often call Staphylococcus epidermidis as a catch-all name), followed by Staphylococcus aureus (13% to 25%) and Streptococcus sp. (18% to 22%, Table 1). In the United States, 21% to 68% of cases of microbial keratitis with microbial growth were caused by gram-negative bacteria, of which the most commonly isolated was Pseudomonas aeruginosa (18% to 93%, Table 1).

A similar spectrum and rate of microbial isolation have been reported from other countries (Lichtinger et al, 2012; Hernandez-Camarena et al, 2015; Prokosch et al, 2012; Young et al, 2013; and others. See Table 1 for full list). Fungi are usually less frequently isolated compared to bacteria, and protozoa (primarily Acanthamoeba sp.) are very rarely isolated (<5%). Differences in the types and rates of isolation probably reflect the penetration of contact lens wear in the market, urban versus rural populations, and climatic differences.

ANTIMICROBIAL RESISTANCE OF BACTERIA ISOLATED FROM MICROBIAL KERATITIS

Several studies have investigated the antimicrobial sensitivity of bacterial isolates from keratitis (Table 2). Examining the types of S. aureus isolates first, we can see that the rate of isolation of methicillin-resistant S. aureus (MRSA) in the United States is approximately 32% of all S. aureus isolates (Ni et al, 2015; Sand et al, 2015; Chang et al, 2015), but this is less in India at approximately 10% of isolates (Lalitha et al, 2016). The worrying issue regarding MRSA is not that it is methicillin resistant, but that this resistance is also accompanied by high rates of resistance to fluoroquinolones and aminoglycosides (Table 2). There is a similar problem with methicillin-resistant Staphylococcus epidermidis (MRCnS; Table 2) (Sand et al, 2015). Some reports do not specifically identify the methicillin-resistant status of isolates of staphylococci, but the rate of resistance to oxacillin is reported (Table 2), and this is essentially the rate of MRSA or MRCnS within that population. Rates of MRCnS have been reported to be as high as 53% of isolates in a report from Mexico, and the rates are often between 40% to 50% of isolates in other countries (Table 2).

For Pseudomonas aeruginosa, isolates from the United States, Canada, Taiwan, Iran, and Saudi Arabia are relatively sensitive to the tested fluoroquinolones and aminoglycosides (Table 2). There are some slight increases in rates of resistance to aminoglycosides in isolates from Mexico and Germany. There is a much higher rate of resistance to fluoroquinolones in isolates from India (Table 2) as well as somewhat higher rates of resistance to aminoglycosides. Given that Park et al (2015) reported that the most popular antibiotic in the United States for treating microbial keratitis of less severe ulcers was moxifloxacin, and the most popular treatment of more severe ulcers was fortified broad spectrum antibiotics (usually aminoglycoside plus cephalosporin), the potential for increasing rates of resistance to these antibiotics is of concern. Continued surveillance and publication of resistance/sensitivity data will inform the choice of antibiotic to be prescribed.

RISK FACTORS FOR CONTACT LENS-RELATED MICROBIAL KERATITIS

Several studies have published since 2011 on the risk factors that are associated with contact lens-related microbial keratitis (Booranapong et al, 2012; Ismail et al, 2015; Lim et al, 2016, and others. See Table 3 for full list). These studies have been undertaken in Australia, France, Singapore, Malaysia, and Thailand. Perhaps unsurprisingly, the risk factors are similar to previous reports. Sleeping in contact lenses is a consistent risk factor in all of the studies, with odd ratios ranging from 1.2 to 6.5 (Table 3). Other risk factors include those relating to maintaining contact lens hygiene (e.g., not washing hands, not rubbing lenses), not replacing lenses as frequently as directed, not replacing contact lens cases frequently (i.e., keeping cases for more than six months), purchasing lenses from the internet or via mail order, and use of multipurpose disinfecting solution (MPDS) compared to hydrogen peroxide (Table 3). Two studies, one in Singapore and one in Australia, identified using a particular MPDS (one with polyhexamethylene biguanide [PHMB] as the only disinfectant) as a major risk factor, with odds ratios of 16.2 and 7.2, respectively (Lim et al, 2016; Stapleton et al, 2012). It is not simply that this MPDS uses only PHMB as a disinfectant, as other PHMB-containing MPDSs are not associated with any increased risk (Willcox and Stapleton, 2014). This highlights the issue of other excipients in MPDSs that influence disinfectant activity.

RISK FACTORS FOR CONTACT LENS-RELATED CORNEAL INFILTRATIVE EVENTS

Corneal infiltrative event (CIE) is a catch-all term to describe infiltration of the cornea without frank infection. It encompasses conditions such as contact lens-induced acute red eye, contact lens-associated corneal ulcers, and symptomatic or asymptomatic infiltrates. CIEs are associated with microbial colonization of contact lenses (Willcox et al, 2011; Szczotka-Flynn et al, 2010), and increased microbial colonization of lenses (also called lens bioburden) is associated with a 5.5- to 8.7-fold increased risk of developing CIEs. Perhaps not surprisingly, given the microbial involvement in CIEs, many of the risk factors for developing these are similar to those for contact lens-related microbial keratitis (Table 4). For example, sleeping in lenses increases risk, as does the age of the contact lens storage case and inappropriate hygiene practices, such as rinsing lenses in tap water rather than a disinfecting solution (Table 4). Use of MPDS compared to hydrogen peroxide for disinfecting contact lenses is a risk for developing CIEs (odds ratios of 2.9 to 18.4, Table 4).

One risk factor that has not been reported for microbial keratitis is an approximate doubling of the risk of developing CIEs when wearing silicone hydrogel lenses compared to traditional hydrogel lenses (Radford et al, 2009; Chalmers et al, 2010; Chalmers et al, 2011; Chalmers et al, 2012). Why this should be is not known, but it could be due to the increased adhesion of bacteria to silicone hydrogel lenses (Henriques et al, 2005; Bruinsma et al, 2001; Willcox et al, 2001).

A risk factor that has been inconsistently reported is the use of daily disposable lenses as compared to daily wear of reusable lenses. Radford et al (2009) reported that daily disposable lens wear was associated with a 2.5-fold increased risk of developing CIEs, whereas Chalmers et al (2012) reported the opposite—that daily wear of reusable lenses was associated with a 12.5-fold risk of developing CIEs compared to daily disposable use. These differences may be due to different usage rates of certain daily disposable lenses, as Radford et al (2009) found that their data was influenced by the use of one particular daily disposable lens type; or perhaps they are due to unreported noncompliance issues with daily disposable use in different studies (such as using lenses for longer than a single day). Richdale et al (2016) have reported that not using the daily disposable replacement schedule increases the risk of CIEs.

THE NORMAL OCULAR MICROBIOTA

Finally, here is some information regarding what constitutes the normal ocular microbiota. First, it’s important to review some vocabulary. I use the term microbiota rather than microflora, as I think this better represents what microbes are: they are small forms of life (micro = small; biota = life); they are not small plants (flora = plants). Microbiota can be defined as the microorganisms of a particular site, habitat, or geological period. Microbes are defined as any type of microorganism (bacteria, fungi, viruses, or protozoa). Microbiology is the science that studies extremely small forms of life. The relatively new term microbiome means the combined genetic material of the microorganisms in a particular environment, similar in concept to the human genome (although the genes in the microbiome vastly outnumbers the genes in our bodies). The microbiome represents the most comprehensive viewpoint of microbes present in a system.

Most studies on the ocular surface microbiota use traditional culture techniques to isolate and identify the microbes. These techniques are still of vital importance in microbiology because, for example, only after growing the microbes can we test whether they are sensitive or resistant to antimicrobial agents. However, we should recognize that we cannot culture all of the microbes in a particular environment. Indeed, it has been estimated that more than 50% of microbial types cannot be isolated by culture; this may be an underestimate in certain environments (Stefani et al, 2015). This notwithstanding, it is still important to understand the culturable microbiota of the normal ocular surface as well as to understand how contact lens wear may affect this culturable microbiota.

Most studies published since 2011 have reported the aerobic microbiota that can be cultured. But, we should keep in mind that using anaerobic or microaerophilic culture conditions can isolate other bacterial types such as the anaerobic Propionibacterium spp. Furthermore, isolation rates can depend on use of appropriate media. These factors may be why rates range in different studies. The aerobic culture of conjunctival swabs continues to show that the most frequently isolated microbial types are bacteria, with coagulase-negative staphylococci being the most common (Dave et al, 2013; Fernández-Rubio et al, 2013; Hsu et al, 2013; Mshangila et al, 2013; and others. See Table 5 for full list). The rate of culture of coagulase-negative staphylococci ranges from 90% to 25% of all swabs processed for culture (Table 5). The next most common aerobic isolate is S. aureus, ranging in frequency from 0% to 11% (Table 5). Other gram-positive bacteria can also be isolated, albeit at usually reduced frequencies (Table 5). Gram-negative aerobic bacteria have also been isolated, with frequencies ranging from approximately 4% to 8% (Table 5).

Some studies have examined the frequency of antibiotic resistance in the normal aerobic culturable ocular microbiota. These studies have reported this microbiota from subjects prior to them undergoing surgery for cataract removal and prior to instillation of prophylactic antibiotics, for example. Thus, these subjects tend to be an older age group. Be that as it may, it is interesting to examine the antibiotic resistance of the normal ocular microbiota, as this may be the source of infecting agents in keratitis and endophthalmitis. The rate of oxacillin resistance (i.e., methicillin resistance) in coagulase-negative staphylococci from conjunctival swabs of these subjects in the United States has been reported to be 48% (Hsu et al, 2013) and from a similar cohort in Uganda to be 32% (Mshangila et al, 2013). For S. aureus, the rate of oxacillin resistance has been reported to be 64% of isolates from the United States (Hsu et al, 2013) and 28% of isolates from Uganda (Mshangila et al, 2013).

Perhaps due to these relatively high rates of oxacillin resistance, the rate of resistance of coagulase-negative staphylococci to ciprofloxacin is 35% in the United States and 24% in Uganda; for S. aureus, it is 55% in the United States and 28% in Uganda (Hsu et al, 2013; Mshangila et al, 2013). In the United States, the rates of resistance to other fluoroquinolones (levofloxacin, moxifloxacin, gatifloxacin; not reported for the Uganda isolates) is also high (35% to 57%) (Hsu et al, 2013). The rates of resistance to gentamicin in isolates from the United States is 5% for coagulase-negative staphylococci and 0% for S. aureus (Hsu et al, 2013), but from Uganda is 21% for coagulase-negative staphylococci and 31% for S. aureus (Mshangila et al, 2013). These relatively high rates of resistance to antibiotics in the normal ocular microbiota is surprising. The fact that the subjects were relatively old may be a factor, as isolation of MRSA increases with age (Nimmo et al, 2007).

A recent study shows the potential importance of the normal ocular microbiota. Kugadas et al (2016) compared the ability of P. aeruginosa to infect the corneas of two types of mice: one was relatively normal (specific pathogen free), and the other was rendered germ-free—that is, there were no culturable microbes on or in them, including on the ocular surface. The authors demonstrated that the germ-free mice were more susceptible to infection by P. aeruginosa, with higher numbers of bacteria being recovered from infected eyes and having a higher clinical response to the P. aeruginosa infection. The authors reconstituted part of the ocular microbiota by instilling coagulase-negative staphylococci onto the ocular surface. The reconstituted mice were better able to control a subsequent infection with P. aeruginosa.

This experiment brings up an important question regarding the development of antimicrobial contact lenses. My laboratory and others have been investigating whether antimicrobial contact lenses can reduce the incidence of CIEs or microbial keratitis. We have shown that antimicrobial contact lenses can reduce the ability of P. aeruginosa to cause both CIEs and microbial keratitis (Cole et al, 2010; Dutta et al, 2016). However, we do not know whether wearing these lenses might also reduce the normal ocular surface microbiota. This will need to be examined in clinical trials of these lenses.

Finally, some information of the microbiome. The microbial genetic material on the ocular surface can be identified from ocular swabs using molecular biology techniques such as the polymerase chain reaction to amplify microbial DNA. The most common technique used is to amplify the 16S rRNA gene of bacteria, which contains elements that are highly conserved between different species of bacteria and so differences in sequences can be used to determine the microbiome of an ecological niche. The use of this technique to evaluate the normal conjunctival microbiome has identified many genera of bacteria that are not grown using traditional culture techniques.

While these microbiome studies show that staphylococci are present on the ocular surface, their relative abundance is apparently much less compared to other microbial types. For example, Huang et al (2016) showed that the most common microbial types in microbiome investigations on the ocular surface were Corynebacterium spp. and Pseudomonas spp. Dong et al (2011) found that Pseudomonas, Bradyrhizobium, Propionibacterium, Acinetobacter, and Corynebacterium spp. were the most abundant microbial types on the conjunctiva using the 16S rRNA technique. Another study that used similar microbiome techniques has shown that contact lens wear can apparently alter the conjunctival microbiome, yielding greater abundance of Methylobacterium, Lactobacillus, Acinetobacter, and Pseudomonas spp. but lower abundance of Haemophilus, Streptococcus, Staphylococcus, and Corynebacterium spp. (Shin et al, 2016). Zhou et al (2014) studied the normal conjunctival microbiome and reported that the most abundant bacteria were Corynebacterium, Simonsiella, and Streptococcus spp.

However, questions do arise from these analyses. Where were the staphylococci that can be cultured from most conjunctival swabs? Why aren’t many of these bacteria cultured from ocular swabs when they are able to be cultured normally? Also, why is the pathogen Pseudomonas present on the normal conjunctiva? The answers may lie in another recent study that used stringent controls and demonstrated that the normal conjunctival microbiome was paucimicrobial and predominantly composed of Corynebacteria, Propionibacteria, and coagulase-negative staphylococci—just like the culturable microbiota (Doan et al, 2016). More work is required to fully understand what constitutes the resident microbiome of the ocular surface and how this relates to the culturable microbiota. CLS

For references, please visit www.clspectrum.com/references and click on document #255.