Patients with corneal infiltrates

This month’s Corneal Health column focuses on identifying cases of infectious infiltrations.

Issue: April 15, 2007

Eric D. Donnenfeld, MD: This case is a little different. This is a patient who is not a contact lens wearer. A 22-year-old woman has acute onset of a corneal infiltrate in the right eye. You can see this classic peripheral infiltration of the cornea (Figure 1). Who thinks this is infectious? Dr. Mah, do you think this is an infection, or do you think this is an immune infiltrate?

Francis S. Mah, MD: Yes, just the pattern looks like it is probably immune with the patient’s blepharitis, meibomian gland dysfunction, all of the blood vessels in the peripheral cornea.

Eric D. Donnenfeld

Dr. Donnenfeld: Do symptoms help you tell if something is infectious or not?

Dr. Mah: Symptoms help, but most people with both processes have a lot of difficulty with pain and so forth. They help separate the two.

Dr. Donnenfeld: Dr. Mah, how would you manage this patient who walks into the office for the first time?

Dr. Mah: Because we have a lab, basically everybody gets cultured with anything that can be cultured. So we would culture this patient. And then we would put the patient on an antibiotic and a steroid. Put her on doxycycline and tell her about some mechanical processes.

Figure 1: Acute onset of a corneal infiltrate
A 22-year-old woman with acute onset of a corneal infiltrate in the right eye.

Images: Ophthalmic Consultants of Long Island

Dr. Donnenfeld: So you would culture the patient and start antibiotic on the first visit on this patient?

Dr. Mah: Yes.

Terrence P. O’Brien, MD: One of the helpful slit lamp biomicroscopic signs is the presence of an intervening clear zone between the limbus and the peripheral corneal infiltrate. That suggests an immune complex deposition, a type III immune mediated reaction where there is a diffusion of the antigen (eg, lipoteichoic acid) into the peripheral cornea. If you have a contiguous spread of infiltrate crossing the limbus, that is usually a type IV delayed hypersensitivity cell-mediated immune reaction. But in that case I think there was an intervening clear zone between limbus and infiltrate. That case was thus probably indicative of staphylococcal hypersensitivity reaction causing the peripheral non-ulcerative infiltrative keratitis. I would agree with Dr. Mah’s excellent approach utilizing culture to be absolutely certain.

Dr. Donnenfeld: Dr. Kim, you were involved in the ASCRS paper on infectious keratitis after LASIK. Which of these patients is worrisome and which is not (Figure 2)? They both had LASIK. They both have inflammation in the interface. Which patient are you concerned about infection and which are you not?

Terry Kim, MD: I would definitely be more concerned with the image on the bottom. The image on the top shows a more diffuse infiltrate consistent with inflammation or diffuse lamellar keratitis (DLK). The image on the right shows a focal infiltrate seated in the interface, and that is the case I would culture immediately.

One point we stress in our ASCRS paper is that whenever you encounter a suspicious infiltrate after LASIK, you should lift the flap and obtain scrapings for stain and culture to rule out infection. As we learned from the ASCRS surveys on this issue, some unusual and opportunistic organisms such as atypical mycobacteria and Nocardia can cause recalcitrant infectious keratitis after LASIK, so identifying these organisms and using appropriate antibiotic therapy is crucial, as opposed to simply starting empiric therapy.

Figure 2a: Diffuse lamellar keratitis Figure 2b: Infectious keratitis
Post-LASIK intralamellar keratitis. Two patient eyes shown with LASIK and infiltration in the interface. DLK (above) and infectious keratitis (below).

Figure 3a: MRSA infiltrate under LASIK flap at presentation Figure 3b: MRSA infiltrate under LASIK flap with flap elevated and scraping stromal bed
MRSA infiltrate under LASIK flap. At presentation (left) and with the flap elevated and scraping the stromal bed for culture (right).

Figure 4a: Vancomycin irrigated under the flap Figure 4b: 4 days later
Initial Treatment. Vancomycin irrigated under the flap (left). Same patient, 4 days later (right).

Dr. Donnenfeld: Any focal infiltrate is important. Here is that same patient being scraped (Figure 3). This patient developed MRSA despite vancomycin therapy around the clock and irrigated in the interface (Figure 4). And we finally ended up removing the flap and applying mitomycin (Figure 5). The patient did pretty well (Figure 6).

I am going to skip through the therapy other than to say that I think you have to scrape every single patient who has an infiltrate after LASIK because of the incidence of opportunistic infections and ones that will not respond to conventional therapy.

Dr. O’Brien: The other slit lamp biomicroscopic sign that is helpful is to look carefully in the anterior chamber for the presence of inflammatory cell, which is typically in response to the infection. If there is no cell present in the aqueous, then the infiltrate is more likely sterile and is usually noninfectious DLK. But I completely agree with Drs. Donnenfeld, Kim and Mah that if there is any doubt or, more correctly, any suspicion that it could be an infection in this setting, then it is mandatory to sample the material for microbial culture.

Figure 5: One week later the necrotic flap was removed
Treatment. One week later the necrotic flap was removed. Topical mitomucin C was applied to prevent scarring.

Figure 6: Weeks later, UCVA was 20/25
Weeks later, UCVA was 20/25

Calvin W. Roberts, MD: And the other thing, too, is organisms. Organisms that have been identified are usually those that are a little bit unique. Most patients these days are on topical fluoroquinolones, so you want to look for things that are not typically covered under topical fluoroquinolones: fungal keratitis, methicillin-resistant infections, mycobacterial.

William B. Trattler, MD: The one thing you can also do is you can see DLK occur years after the surgery. We have seen a number of cases in our practice where patients have had corneal abrasions or any type of epithelial defect and even peripheral corneal ulcers that can lead to inflammatory cells depositing in the interface. So late DLK is something to watch for even years later.

For more information:

Eric D. Donnenfeld, MD, is a cornea specialist in private practice at Ophthalmic Consultants of Long Island and co-chairman of Cornea and External Disease at Manhattan Eye, Ear and Throat Hospital. He can be reached at Ryan Medical Arts Building, 2000 North Village Ave., Rockville Centre, NY 11570; 516-766-2519; fax: 516-766-3714; e-mail: eddoph@aol.com.

Terry Kim, MD, can be reached at Duke University Eye Center, Erwin Road, P.O. Box 3802, Durham, NC 27710-3802; 919-681-3568; fax: 919-681-7661; e-mail: terry.kim@duke.edu.

Francis S. Mah, MD, can be reached at University of Pittsburgh Medical Center, Eye and Ear Institute, 203 Lothrop St., 8th Floor, Pittsburgh, PA 15213; 412-647-2200; fax: 412-647-5119; e-mail: mahfs@upmc.edu.

Terrence P. O’Brien MD, can be reached at Bascom Palmer Eye Institute, 7108 Fairway Drive, Palm Beach Gardens, FL 33418; 561-515-1544; fax: 561-515-1588; e-mail: tobrien@med.miami.edu.

Calvin W. Roberts, MD, can be reached at 876 Park Ave., New York, NY 10021; 212-734-7788; fax: 212-734-4476; e-mail: robertsmd1@aol.com.

William B. Trattler, MD, can be reached at 8940 N. Kendall Drive, #400, Miami, FL 33176; 305-598-2020; fax: 305-274-0426; e-mail: wtrattler@earthlink.net.