One recent Saturday morning, I found myself sitting in clinic looking at the perfect microcosm of the controversies inherent in corneal ulcer management. The rather non-descript ulcer visible in the slit lamp had been previously cultured and started on broad-spectrum dual antibiotic therapy by a colleague.

At the time, the pathology appeared to respond favorably to therapy. And yet, for some reason, here was the patient—in clinic on a Saturday, with worsening symptoms and only marginally changed signs.

This prompted at least two immediate questions:

(1) Did these changes represent worsening of an infectious process?

(2) Considering that the patient was already presumably on maximum antibiotic therapy (a fourth-generation fluoroquinolone and fortified gentamicin), what else could be done?

With infectious corneal ulcers—some of the most demanding clinical entities in eye care to treat—these dilemmas are par for the course. Microbial keratitis (MK) challenges your clinical decision making skills, diagnostic skills, patient communication skills and therapeutic skills.

Patients presenting with MK require a flexible approach to treatment—not the simple, cookie-cutter methods we like so much. Within the realm of infectious corneal ulcers, inflexible thinking, blind commitment to diagnoses and inability to react to unexpected results can lead to poor patient outcomes—namely, the prospect of your patient’s eye ending up in a jar, sitting in a lab.

On top of the intrinsic challenges posed by MK, there are significant gray zones within this niche of eye care, e.g., the role of microbiologic studies and the use of steroids in the peri-infectious period. Because the stakes are high and there is so much debate, I present the following discussion as a basis to understand some of these controversies. I will also discuss how to apply a variety of techniques to help establish your own comfort level when managing these challenging cases.

When Should You Culture?
There are three different mindsets to this foundational question in the management of MK: (1) always culture, (2) always treat empirically prior to culture, and (3) only culture in select cases. Of these three approaches, I fall in the case-by-case approach.

I feel that always treating empirically is not an entirely realistic option. When determining your own philosophy, it is useful to look at the strengths and weaknesses of each approach, and understand how those decisions impact the corneal response to infection. Culturing corneal ulcers provides some significant benefits that clarify the approach to treatment.

First—and perhaps most obviously—culturing, when performed appropriately, may provide diagnostic and treatment information that would otherwise be absent from the case. From corneal smears alone (i.e., gram or geimsa stains), you can gather near immediate information about the infectious etiology, which will impact treatment. Contrary to what you may think, gram stains don’t just identify gram-positive or negative organisms; they can also be used to identify acid-fast organisms and fungal elements.

When cultures can be taken with agars and/or nutrient broths, identification of the actual organism may be achieved. Sensitivity and susceptibility testing—valuable in determining the most effective treatments—may also be possible with this method. An infiltrate that is localized, with minimal surrounding edema, round or oval with relatively defined, minimally necrotic margins and an epithelial defect could likely be thought to be a staphylococcal or streptococcal MK.

In the past, empiric treatment consisting of a fluoroquinolone would have been appropriate. However, with the increasing trend toward multiple drug resistance seen in both Staph. and Strep. species, it would be reasonable to anticipate that culture-generated sensitivity and susceptibility testing may play a more prominent role in our treatment strategies. At the very least, it should be considered as an option—even when a reasonable guess at the presumed etiology can be made.

The next benefit of culture is not diagnostic but therapeutic. By removing easily debrided infected tissue, the clinician is essentially debulking the infectious load. Small superficial infiltrates can be grossly removed in their entirety. Further, built-up necrotic tissue impedes corneal healing, so removing this tissue has some benefit in hastening wound healing, as well as enhancing penetration of antimicrobials.

Given both the therapeutic and diagnostic benefits offered by culturing, what possible incentive could one have to avoid it? Well—inertia is a funny thing, and overcoming it can be quite difficult. If you don’t routinely culture, you may not have on hand the necessary prerequisites: media, swabs, slides and a relationship with a microbiology lab. This can make the prospect of culturing seem an excessively time consuming and intimidating process. This alone may be a sufficient deterrent in borderline cases to simply avoid a culture.

I would point out, however, that while this approach may be easiest for you as the clinician, it is not always in the best interest of the patient. So, in cases of large or central ulcers, or those with irregular infiltrates—for example, those with feathery margins, satellite lesions, perineural infiltrates or an amorphous shape—where culturing material, skill or processing facilities are unavailable, referral for these services is mandatory.

Culturing materials
The last drawback to culturing isn’t really a drawback at all, but a reminder to view cultures for what they are worth, and to keep an open mind—especially when dealing with MK. A negative culture may lead to confusion among clinicians; often, they are unsure whether they are still dealing with an infectious etiology. Keep in mind, however, that cultures only return positive in approximately 40-65% of cases of presumed MK.^1,12 Therefore, when compared to many other available diagnostic tests, a culture has relatively low sensitivity, and may need to be repeated as clinical response dictates.

At the very least, negative cultures should not be viewed as evidence of non-infectious keratitis. Similarly, positive cultures that do not fit the overall clinical picture may be reasonably interpreted with scrutiny; consider the possiblility of a contaminated result.

So, how does one decide which infiltrate should be cultured and which to treat empirically? At our clinic, we typically let the infiltrate guide culturing practice. This method is primarily based on size, depth and location. We give preference to culturing larger and more deeply ulcerated lesions, as well as those within close proximity to the visual axis.

I tend to culture infiltrates larger than 2mm in any meridian as measured with a slit-lamp reticule, assuming they are in the mid periphery. For smaller infiltrates, I will typically culture if they are more centrally located. As with most aspects of eye care, however, the criteria for culturing aren’t always black and white. Other factors may sway me towards more aggressive culturing.

Obviously unusual elements in the patient history—such as recent vegetative trauma, nosocomial-based infection and infections in immuncompromised patients (both in cases of systemic immune deficiencies and in solely ocular cases as with corticosteroid-treated eyes post-corneal transplants)—may increase the likelihood that I perform a culture.

Also, a great deal of information regarding a differential diagnosis can be obtained by closely scrutinizing the infiltrate. In cases where the infiltrate exhibits feathery margins, an arborized pattern, suppurative necrotic margins or a granular appearance, I would be more likely to culture using specific media.

On the other hand, I feel more comfortable treating patients empirically when they present with small and peripheral infiltrates with regular features. This practice is generally backed up by literature that shows a relatively good success rate with empiric strategy for mild corneal ulcers; however, for severe cases treated empirically, the outcomes were significantly worse.^1,2
    
Broad-spectrum Antibiotics vs. Fortified Topicals
There is a mini-controversy loosely tied to culturing practice in management of MK: what’s the role of commercially available broad-spectrum antibiotics as monotherapy in empiric treatment strategies? The current iteration of the question compares the use of fourth-generation fluoroquinolones, moxifloxacin and gatifloxacin to fortified antibiotic alternatives.

For young doctors such as myself, who have only practiced in the era of effective broad-spectrum commercially available antibiotics, there may be some confusion as to why this question is even being asked.

	Table 1. Principles for Successful Use of Corticosteroids
	(1) Perform scrapings for stain and culture.
	(2) Use adequately dosed bactericidal antibiotics.
	(3) Delay initiation of steroids until a clearly beneficial effect to antibiotic has been determined.
	(4) Continue concurrent use of antibiotic with steroids.
	(5) Delay use of steroids if the causative organism is not identified. Use 2-5 days after appropriate antibiotic therapy.

To clarify, a bit of history is in order. Prior to the development and dissemination of good, broad-spectrum topical antibiotics—namely fluoroquinolones—the first-line treatment of bacterial corneal ulcers was paired fortified antibiotics. As the well-tolerated, widely effective fluoroquinolone class came to market, community-based eye care practices transitioned to these readily available agents in lieu of fortified agents.

Now, the question of their propriety is being raised again, as strains of bacteria resistant to fluoroquinolones, including the fourth-generation group, have become increasingly common.

Currently, research suggests that fourth-generation fluoroquinolones continue to work as effectively in the empiric treatment of non-severe MK as paired fortified antibiotics.^1,3-5 However, this research should be interpreted with a caveat. As stated, there has been a generalized trend towards resistance to fluoroquinolones, with older generations being the most impacted. Moxifloxacin and gatifloxacin have encountered resistance as well.

One 15-year review of the antibiotic sensitivities of coagulase positive staphylococcal isolates causing endophthalmitis clearly illustrates this shift. In the first five years of the study, the researchers concluded the sensitivity of these isolates to moxifloxacin and gatifloxacin to be over 90%.⁶ However, in the most recent five years, these values had dropped to less than 75%.⁶

Besifloxacin would be an obvious alternative to these medications. It is a novel fluoroquinolone often described as a fourth-generation medication, but with molecular formula more akin to that of a second-generation drug. It is used exclusively in an ocular formulation; by obviating systemic exposure, it could be expected to encounter less resistance.

Unfortunately, this is a slightly flawed argument. Any bacteria that are resistant to the true fourth-generation medications should also exhibit resistance to besifloxacin. 

When studied in vitro against variably resistant organisms, however, besifloxacin seems to be just as effective—and in some cases more effective—than gatifloxacin or moxifloxacin.  Regardless, as time passes and resistant strains develop, the balance may once again tip in favor of fortified drugs. Current practice for empiric treatment at our facility is to supplement a newer generation fluoroquinolone—usually besifloxacin—with one fortified antibiotic.

Role of Steroids
The last question that comes up from time to time is the appropriateness of corticosteroids in the treatment of bacterial ulcers. The Steroids for Corneal Ulcer Trial (SCUT) group most recently attempted to address this question. While the SCUT study was large and well designed, its clinical application is difficult to determine.

Primary and secondary outcome measures among the groups treated with placebo, as well as those treated with a corticosteroid, were not significant. So, according to research conducted by SCUT, there is no benefit or drawback to the use of topical steroids when used in the treatment of bacterial corneal ulcers.

To put it in a different way, there was no significant difference in visual outcome, scar size, time to re-epithelialization or progression to perforation between the two groups. However, in subgroup analysis, there was a modest benefit in three-month visual acuity outcome in steroid-treated patients when the ulcers were central or deep.

Patients presenting with best spectacle correction of counting fingers or worse also exhibited a modest benefit from corticosteroid treatment. The upshot: those patients with the worst ulcers at presentation may experience some benefit from corticosteroid use. The authors of the study rightly note that this patient population, with the most to gain per study results, is also the group for whom managing clinicians will feel least comfortable prescribing steroids.⁹

One critical point to understand about SCUT is that it only studied cases of bacterial keratitis. Therefore, parallels should not be drawn between its results and the value of steroids for either non-bacterial or atypical bacterial (e.g., Nocardia, Mycobacterium) etiologies of MK. In cases of bacterial corneal ulceration, if you feel steroids must be used, it is probably useful to draw a parallel to the initial treatment period of corneal ulcers with the timeframe after instituting corticosteroids.

In the initial period of corneal ulcer management, the patient is often seen daily for several visits. This is done to establish whether there has been a positive treatment response. Once a patient is started on steroid treatment, it is probably wise to question whether the change you are making to treatment will alter the success of the antimicrobial strategy previously applied.

Within our facility, corticosteroids are used periodically as adjunctive agents for some infectious ulcers. Prior to their use, there is always a clear picture of improving clinical signs indicating treatment success. Further, steroids are often introduced in a strict trial fashion.

Jeff Urness, a former colleague and mentor of mine, impressed upon me the value of a steroid trial consisting of a single drop in the office and then a close follow-up.

Self-Assessment
When confronted with a corneal ulcer, you first need to ask yourself to gauge your comfort level with the particular case. A typical battery of questions during this analysis might include:

• What is the size, depth and location of the infiltrate?

• Are there other irregular features associated with the case? For example—does the patient have a hypoyon or history of vegetative trauma preceding the ulceration?

• Do you feel it needs to be cultured? If so, are you competent in this skill?

• Do you have ready access to a lab facility and a good working relationship with the professionals there?

• Are you both able and willing to follow-up closely and to observe closely for signs of improvement or worsening?

Unfortunately, as we all know, ulcers do not respect typical business hours. If you feel comfortable after answering these questions, it is appropriate to take on the case in question.

However, that is only the beginning of corneal ulcer treatment. Each subsequent visit requires close evaluation for treatment response; if the case does not proceed as expected, something needs to change. The necessary modification will depend on the specific case.

You may need to attempt a culture after not previously performing one, or you may decide to perform a second culture—perhaps with a 24-hour washout period where antibiotics are not used—in the face of a negative result. In some scenarios, it may be best to refer the patient out. The biggest concern that may face a clinician considering taking on a role in the treatment of corneal ulcer is one of liability.

Practice patterns will necessarily be individualized based on factors such as one’s training, breadth of clinical experience and ability to consult swiftly with corneal specialists. As an optometrist in a multidisciplinary setting, I have a consultation with the specialist in nearly any situation in which there is a question regarding an atypical ulceration or consideration of a potentially controversial treatment regimen change.

More often than not, we are in agreement; however, it’s always nice to have his signature in the chart accompanying any of these changes. If I were to go out and practice on my own, I doubt that I would have the stomach for the liability that accompanies the management of any unusual or severe corneal infection, despite being quite comfortable with my skill in these cases.

Ultimately, the decision to treat or refer has to do with each individual doctor’s comfort with the clinical case and their tolerance for the liability involved.

So, while the management of corneal ulcers may seem daunting (it can be) or challenging (it is), these cases present an opportunity to practice at the highest level of the profession. They also allow us the opportunity to offer a great service to patients who are in need of quality care. In my experience, I’ve not found any other result in eye care to be as satisfying as a corneal ulcer with a good outcome.

As for my weekend case described at the outset of this article, I decided to re-culture her ulcer. Of course, it came back fungal on interpretation of smears. Now—where do I get fortified voriconazole on a Saturday?

	Joining the Culture Club
	Taking the steps necessary to add a basic culturing component to your practice can seem daunting—both from a clinical standpoint as well as a logistical one. But, if you plan to treat corneal ulcers, a culturing component should be something that you have in place. A proper culturing practice needs the following items: a small refrigerator to store culture material, handheld implements with which to take a culture, and media and slides to inoculate the material onto. If you plan to read the gram strain results yourself, you’ll also need microbiologic stains and a microscope. The practice will also need to establish the following features: a courier service with a microbiology lab (if self-transport is not feasible), a way to re-supply fresh culture material (as agars have a finite life span) and, of course, access to a microbiology lab to process cultures. In my experience, these labs have been both enthusiastic and helpful in supplying media and in giving direction for best practices in processing the media. Typically, these microbiology labs can often be found as part of local hospitals, though stand-alone labs also exist. Acquisition of tools such as a platinum Kimura spatula and burner for sterilization can be acquired through medical supply catalogs. One last item that may prove useful is a good reference text. I’ve found Haesaert’s Clinical Manual of Ocular Microbiology and Cytology to be a great, relatively inexpensive aid. This text provides a detailed description of culturing techniques and conventions. When performing the actual culturing process, a spatula should be used while preparing microscopic slides, but a calcium agininate swab—when appropriately used—can produce yields that are as good or better than a spatula for inoculating media. Also remember: the zone of the infiltrate with the most recoverable micro-organisms tends to be at the lesion’s edge rather than center.10,11

1. Mcleod SD, et al. The Importance of Initial Management in the Treatment of Severe Infectious Corneal Ulcers. Ophthalmology. 1995; 102: 1943-1948.
2. Rodman RC, et al. The utility of culturing corneal ulcers in a tertiary referral center versus a general ophthalmology clinic. Ophthalmology. 1997;104: 1897-1901.
3. Sharma N, et al. Evaluation of moxifloxacin 0.5% in treatment of nonperforated bacterial corneal ulcers: a randomized controlled trial. Ophthalmology. 2013;120: 1173-1178.
4. Shah VM, et al. Randomized clinical study for comparative evaluation of fourth-generation fluoroquinolones with combination of fortified antibiotics in the treatment of bacterial corneal ulcers. Cornea. 2010;29: 751-757.
5. Hanet MS, Jamart J, Chaves AP. Fluoroquinolones or fortified antibiotics for treating bacterial keratitis: systematic review and meta-analysis of comparative studies. Can J Ophthalmol. 2012;47: 493-499.
6. Miller D, et al. In vitro fluoroquinolones resistance in staphylococcal endophthalmolmitis isolates. Arch Ophthalmol. 2006;124: 479-483.
7. Sanders ME et al. Efficacy of besifloxacin in a rabbit model of methicillin-resisitant Staphylococcus aureus keratitis. Cornea. 2009;1055-1060.
8. Sander ME, et al. Comparison of besifloxacin, gatifloxacin and moxifloxacin against strains of pseudomonas aeruginosa with different quinolone susceptibility patterns in a rabbit model of keratitis. Cornea. 2011;30: 83-90.
9. Srinivasa M et al. Corticosteroids for Bacterial Keratitis: The Steroids for Corneal Ulcers Trial (SCUT). Arch Ophthalmol. 2012;130: doi: 10.10001/archophthalmol.2011.315
10. Benson WH, Lanier JD. Current diagnosis and treatment of corneal ulcers. Curr Opin Ophthalmol. 1998;9: 45-49.
11. Jacob P et al. Calcium algninate swab versus Bard Parker blade in the diagnosis of microbial keratitis. Cornea. 1995;14: 360-364.
12. Levey SB  et al.  The Role of Cultures in the Management of Ulcerative Keratitis.Cornea. 1997;16: 383-386.