Cornea Health Round Table: Benefits vs. risks of simultaneous cross-linking and PRK

A nonhealing epithelial defect after PRK or cross-linking can be considered a medical emergency because of the increased risk for visually significant corneal scarring that may develop. Indeed, there are many similarities between PRK and cross-linking. Not only do they share similar side effects, but also their preoperative and postoperative managements are similar, according to Eric D. Donnenfeld, MD.

In this issue of Ocular Surgery News, Donnenfeld and a panel of experts convene for the annual Cornea Health Round Table at OSN New York 2017 where they examine a complicated case of persistent epithelial defect after simultaneous PRK and cross-linking, as well as a case of an infected arcuate incision.

Donnenfeld: Cross-linking was approved in 2016 in the United States, and we were very happy about that. It is basically the application of UV light and topical riboflavin, which is a B vitamin. Preoperative and postoperative management of both cross-linking and PRK are very similar — bandage contact lens, fluoroquinolone antibiotic, corticosteroid and topical NSAIDs.

Richard L. Lindstrom, MD: There is kind of a raging controversy that we could discuss about whether you should do PRK and then cross-link immediately or whether you should cross-link and then do PRK later.

Donnenfeld: A lot of people very strongly feel that you do the PRK, you remove the tissue, and then you cross-link because you are cross-linking the tissue after you remove it. It does not make sense to cross-link the tissue that you are going to remove 3 months later. On the other hand, it adds two insults: UV light exposure and toxicity of the riboflavin and the laser. The concern is that you are going to have aberrant healing. So you are weighing the aberrant healing vs. the better cross-linking.

Source: Barbara Galati

A. John Kanellopoulos, MD, who is, in my opinion, the guru of cross-linking, recommends almost always doing them at the same time. I have done that about 20 times now, and I have been burned by patients having poor healing. So now, 99% of the time, I do the cross-linking and 3 months later I do topographic laser.

William B. Trattler, MD: I agree 100% with performing sequential CXL followed months or years later by PRK, rather than combining the procedures. I performed a combined treatment in a patient, and the patient experienced a significant delay in epithelial healing, as well as significant corneal haze. Thankfully, the corneal haze improved over time, but she remains with loss of BCVA due to the residual corneal haze and irregular astigmatism. She currently wears a scleral lens. My opinion is that we should perform either cross-linking or PRK first, as there can be increased risk for a poor visual outcome when the two procedures are combined.

Donnenfeld: How long do you wait?

Trattler: If cross-linking is performed first, the longer the patient waits, the greater the improvement in corneal shape. At 3 months after cross-linking, the corneal shape is typically about the same. At 6 months, there is often improvement in corneal shape. At 1 year, there is more flattening/improvement than at 6 months, so the amount of laser reshaping that is required is less than if PRK is performed at 3 months. In my practice, I have patients wait a year or longer and then do the PRK procedure.

John P. Berdahl, MD: I have a couple of comments. First, with the emergence of cross-linking, LASIK becomes safer. The reason why you did not do LASIK in the first place on that patient was for risk of ectasia because previously we did not have solutions for it. Now we do. Second, I think that excimer and particularly topo-guided treatments over these aberrated corneas, whether they are prior LASIK, prior scar or keratoconus, are going to be a huge unmet need that we fill for people, and I just hope that it gets easy enough and standardized enough that we can do it with confidence.

Donnenfeld: When patients come in now for cataract surgery with obvious ectasia, keratoconus or even forme fruste, we can do cataract surgery and probably put in a toric lens, but putting a toric lens in a keratoconic eye is like putting a square peg in a round hole. It is never going to fit exactly right. So I ask the patient, “If you are willing to spend the time with me, we’ll cross-link you, we’ll do a topographic laser, we’ll bring you back 6 months after that and then do your cataract surgery.” Then we are going to really change their lives because we are going to give them real quality vision. You cannot give somebody with an irregular cornea quality of vision with glasses — it just cannot happen. If they are willing to wear a gas permeable contact lens, that is a different story.

Berdahl: How have your results been, and how difficult is it to make the calculations?

Donnenfeld: Because we are doing fairly neutral ablations on these patients, they are much more precise and much more exact after we do this because the corneas are more regular. Getting the IOL correct in a keratoconic eye is very difficult.

Berdahl: How hard is it to plan the laser ablation?

Donnenfeld: It is typical laser ablation. We are not treating refractive error at all. We are just treating irregularities. We do a zero-sum topographic laser and make the cornea more regular.

Case one

Donnenfeld: A 55-year-old man with keratoconus who is contact lens intolerant undergoes topographic PRK and simultaneous riboflavin UV cross-linking. A bandage contact lens is placed, and 5 days later he has a persistent epithelial defect. It is just not healing (Figure 1). Now I am worried. This to me is a serious problem. This is not a minor problem — this is a refractive emergency. How would you handle this?

Penny A. Asbell, MD, FACS, MBA: How many benzalkonium chloride (BAK) preserved drops is the patient on? I would definitely want to decrease that if possible, and I would agree with maybe an amniotic membrane that you would do in the office. I would also consider a temporary tarsorrhaphy or lid closure.

Lindstrom: You learn sometimes from clinical trials. We were involved in a clinical trial for the treatment of persistent epithelial defect. The protocol called for stopping all medications and using non-preserved tears, washout and then randomizing to diluent vs. drug. One hundred percent of them healed. We had no one, not one patient, to enroll in the study because when we stopped all meds and put them on non-preserved tears only, they all healed. So that is what I do now: a good bandage contact lens, a Botox (botulinum toxin type A, Allergan) tarsorrhaphy where we inject a little Botox so the lid droops down, and then just non-preserved tears. Almost all of them heal.

Source: Eric D. Donnenfeld, MD

Donnenfeld: Would you stop antibiotics on a patient with an epithelial defect like this (Figure 1)?

Lindstrom: I would be OK with something benign, but I pretty much have not seen them get infections. You have to watch them carefully, but they heal.

Donnenfeld: Penny’s comment was a good one. Stopping the medication with BAK will increase the chance the epithelium will heal normally. There is one antibiotic that is commercially available that has no preservative in it.

Lindstrom: Vigamox (moxifloxacin, Alcon/Novartis).

Donnenfeld: That might be a good antibiotic to use under these circumstances. We stopped the medications, we put him on oral doxycycline, we placed a bandage contact lens. When this did not heal, we then placed an amniotic membrane Prokera (Bio-Tissue). The reason the patient was not healing was he had floppy eyelid syndrome. It is the single most common lid abnormality associated with keratoconus, and it is the reason why the patient was not healing. You can use all the autologous serum or all the tears you want, but this is not going to heal. How would you manage this?

Asbell: Closing the lid with a temporary tarsorrhaphy is very easy to do in the office — no cutting, just use a suture. I numb up the lid a little bit and use 5-0 nylon. You can put it in while the patient is in a slit lamp room; you do not even need a minor operating room.

Donnenfeld: So you can do a suture tarsorrhaphy or use a Botox tarsorrhaphy.

Asbell: I would not use Botox here because it is not going to give you the closure you need.

Lindstrom: Right. Not Botox for floppy lid, but maybe use tissue adhesive.

Donnenfeld: We do glue tarsorrhaphies all the time now. It is a great little procedure. Medical grade cyanoacrylate is applied to the lashes, the patient closes his eyes, and it lasts for about a week. This has been a game changer for us. It is a 1-minute procedure that can save significant vision. To remove the glue is a little bit troublesome sometimes. The key is to cut off the eyelashes, which some patients do not really like.

Lindstrom: For some patients, we use tape for a few days to see if that is enough.

Donnenfeld: We sometimes have them put on a shield when they sleep, but they are easily knocked off during the night.

Lid management

Donnenfeld: So this comes up time and time again. Where do the organisms that cause endophthalmitis come from most commonly? Do they come from the surgeon’s hands? From contamination of the irrigating solution, the patient’s skin or the patient’s lid margin?

It is all about the lid margins, which is why, time and time again, we treat the lid margins. What are your pearls for lid margin management before cataract surgery?

Trattler: When I evaluate patients before cataract surgery and I look at the eyelids, if they have blepharitis, I am a big fan of hypochlorous acid. It is a very powerful antiseptic that can reduce that bacterial load. That is what I often start on my patients before cataract surgery.

Asbell: The key is to look at those lids and assess them and treat them before surgery.

Donnenfeld: You think of methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis as being rare organisms, but the bottom line is that about half of patients coming in for routine cataract surgery today have MRSA or MRSE. So when you go into the operating room, you assume every single patient you operate on has a methicillin-resistant species. It is just extraordinarily common. Almost 50% of patients undergoing routine cataract surgery are colonized with methicillin-resistant staph.

Penny, from your work on the ARMOR study, tell us about besifloxacin.

Asbell: Besifloxacin is not used systemically, so we do not have systemic break points like we do with other antibiotics. But looking at the minimum inhibitory concentration, which is certainly the laboratory test for potency, it matches up very well competitively with vancomycin, which of course has to be compounded. So, you have a drug that you can get off-the-shelf vs. something you might have difficulty obtaining.

Donnenfeld: The other thing is good surgical technique. If you lose vitreous, you will have a rupture of the posterior capsule, and your risk of endophthalmitis skyrockets. You have to be aware of that. Wound leakage is obviously very important as well.

Getting back to what Bill said, reducing preoperative ocular surface flora reduces the risk of endophthalmitis. Betadine scrubs are really good, but hypochlorous acid might be even better in some ways.

Francis S. Mah, MD: Avenova (NovaBay) is the most popular, but there are others. Ocusoft has a product. It does really kill much more rapidly than Betadine (povidone-iodine, Purdue Pharma). Betadine has to kill on contact, and you have to wait for it to kill. Hypochlorous acid is also much less toxic. It is almost like it was made for the eyelids.

Donnenfeld: I think it is a great preoperative prophylaxis for cataract, PRK and LASIK as well.

Mah: You want to put it right at the base of the eyelashes. Demodex, which can be a component of blepharitis, resides in the sleeves and in the follicles, so you want to apply the hypochlorous acid right to the base of the eyelashes.

Trattler: I have my patients spray Avenova on their fingertip and then apply it directly to their eyelashes and their eyelid. It takes them a second or two, so this helps with compliance.

Asbell: What happens when they spray it in their eyes?

Trattler: In my experience, Avenova is comfortable when the solution is exposed to the eye. No burning. I demonstrate on my own eyelids about 8 to 10 times a day at my office, and I never experience irritation.

Lindstrom: A lot of patients have mild acne, rosacea and the like, so I tell them to spray it on the skin and brows. It is nontoxic and has a mild anti-inflammatory effect.

Asbell: So do you use it in the OR before the case? Do you spray it on the patient instead of Betadine?

Lindstrom: Not as a routine, but occasionally when a patient has a history in the OR with povidone, that is what I use.

Donnenfeld: Consider it as an alternative. Betadine is great; it is relatively nontoxic, but it does burn and irritate a little bit. For retina surgeons who are doing intravitreal injections and for our surgery, I think this is a better Betadine. It does everything Betadine does, but does it less toxic, more rapidly and with a higher kill rate. So just put it on the back burner as something you might want to consider adding to your practice.

Next case

Donnenfeld: This is a case that is ongoing. A 58-year-old man had uneventful laser cataract surgery with arcuate incisions. He flew in for the surgery and flew out the next day. Ten weeks later, he presented with an infection — pain and decreased vision (Figure 2). Gram stain was negative. I put him on fortified antibiotics, like we all would do, and he continued to get worse. It is probably fungal, so we added voriconazole, which I almost never do, just to be safe. At about 2 weeks, we cultured again. We stopped all the medications for 36 hours. We re-cultured and still negative. So then I turned to Hank Perry for help.

Henry D. Perry, MD: I was very suspicious. He had three cultures, and everything was read as negative. I stopped all medications for 48 hours and then used an air injection needle in the relaxing incision and scraped the edges and sent the smear for Gram stain, as I suspected microsporidiosis. The pathologist called me an hour later and said it was negative. I just could not believe it. I went to the lab myself. When you really take a look at it, you see a lot of stuff in those epithelial cells. They look like they are almost going to vomit little blue bodies. And these little blue bodies are actually microsporidia, an obligate intracellular parasite, and that is why all the cultures were negative. It is not going to grow on any culture medium. Microsporidiosis typically occurs as a conjunctivitis in HIV patients but is a rare cause of corneal infections in immunocompetent patients.

So the problem was, what to do now? We had known about albendazole as a very effective oral treatment. However, topical Fumagilin-B in most studies is the gold standard. Fortunately, it is commonly used by veterinarians for bees (I ordered it on Amazon), and we used 3 mg/mL. We kept him also on the voriconazole. Interestingly, microsporidiosis is now being reclassified as a fungus. It was a very difficult case, the therapy is long at 6 to 8 weeks, but he is finally doing better. When he presented, he had a hypopyon and counting fingers vision, and now he is at 20/80. He is doing pretty well, but he still has a ways to go.

• References:
• Asbell PA, et al. J Cataract Refract Surg. 2016;doi:10.1016/j.jcrs.2016.11.008.
• Haas W, et al. Am J Ophthalmol. 2011;doi:10.1016/j.ajo.2011.03.010.
• Speaker MG, et al. Ophthalmology. 1991;doi:10.1016/S0161-6420(91)32239-5.

• For more information:
• Penny A. Asbell, MD, FACS, MBA, can be reached at Mount Sinai Medical Center, Box 1183, New York, NY 10029; email: penny.asbell@mssm.edu.
• John P. Berdahl, MD, can be reached at Vance Thompson Vision, 3101 W. 57th St., Sioux Falls, SD 57108; email: john.berdahl@vancethompsonvision.com.
• Eric D. Donnenfeld, MD, can be reached at Ophthalmic Consultants of Long Island, 711 Stewart Ave., Suite 160, Garden City, NY 11530; email: ericdonnenfeld@gmail.com.
• Richard L. Lindstrom, MD, can be reached at Minnesota Eye Consultants, 9801 Dupont Ave. South, Suite 200, Bloomington, MN 55431-3200; email: rllindstrom@mneye.com.
• Francis S. Mah, MD, can be reached at Scripps Clinic, 10666 N. Torrey Pines Road, MS 214, La Jolla, CA 92037; email: mah.francis@scrippshealth.org.
• Henry D. Perry, MD, can be reached at Ophthalmic Consultants of Long Island, 2000 N. Village Ave., Rockville Centre, NY 11570; email: hankcornea@gmail.com.
• William B. Trattler, MD, can be reached at Baptist Medical Arts Building, 8940 N. Kendall Drive, Suite 400-E, Miami, FL 33176; email: wtrattler@gmail.com.

Disclosures: Asbell and Berdahl report no relevant financial disclosures. Donnenfeld reports financial disclosures for Allergan, Alcon, Bausch + Lomb and Johnson & Johnson. Lindstrom reports relevant financial disclosures for NovaBay, Alcon/Novartis, Allergan, Bausch + Lomb, Sun Pharma and Imprimis. Mah reports conflicts of interest with Alcon, Allergan, Bausch + Lomb and Sun. Perry reports financial disclosures for Alcon, Allergan, Bausch + Lomb, BlephEx, NovaBay, Omeros and PRN. Trattler reports he is a consultant and speaker for Allergan, Bausch + Lomb, NovaBay, Johnson & Johnson and Avedro; is a consultant to Alcon; and has a financial interest in CXLO.