Retinal Surgery and Corneal Defects

A joint subspecialty roundtable.


Despite changes in surgical technique that reduce risk, retina specialists still encounter corneal defects with vitrectomy and intravitreal injections. Retinal Physician editor in chief Peter K. Kaiser, MD, leads a discussion with retina specialists Carl D. Regillo, MD, FACS, and Thomas A. Albini, MD, and cornea specialists Marguerite McDonald, MD, FACS, and Francis S. Mah, MD, to try to reach consensus on best practices, tips, and techniques to help prevent and manage corneal defects and persistent corneal defects after vitreoretinal surgery.

Peter K. Kaiser, MD, is the Chaney Family Endowed Chair in Ophthalmology Research and a professor of ophthalmology at the Cole Eye Institute, Cleveland Clinic Lerner College of Medicine in Cleveland, Ohio. Carl D. Regillo, MD, FACS, is chief of the retina service at Wills Eye Hospital and a professor of ophthalmology at the Thomas Jefferson University in Philadelphia, Pennsylvania. Thomas A. Albini, MD, is an associate professor of clinical ophthalmology at Bascom Palmer Eye Institute in Miami, Florida. Marguerite McDonald, MD, FACS, is a refractive, cornea, and cataract surgeon with Ophthalmic Consultants of Long Island in New York and a clinical professor of ophthalmology at Tulane University and New York University. Francis S. Mah, MD, is director of cornea and external disease and co-director of refractive surgery at Scripps Clinic Medical Group in La Jolla, California. Dr. Regillo reports research grant support from Alcon, Allergan, Bayer, Genentech, Novartis, Iconic Therapeutics, and Regeneron as well as consultancy to Alcon, Allergan, Genentech, Novartis, Iconic Therapeutics, Shire, and Notal Vision. Dr. Albini reports no related disclosures. Dr. McDonald reports consultancy to Bio-Tissue, Allergan, Shire, BlephEx, and J&J Vision. Dr. Mah reports consultancy to Bausch + Lomb, Novarits, and Allergan.

Dr. Kaiser: Thank you all for joining the roundtable. Dr. Regillo, how often do you see corneal epithelial defects in the perisurgical period, when do you see them, and what sort of risk factors do you think lead to them?

Dr. Regillo: Potential corneal issues, mainly epithelial defects including abrasions, can be seen after vitrectomy and after office-based intravitreal injections. Fortunately, we see fewer postsurgical corneal problems than we used to. Much of this can be attributed to advances in vitreoretinal surgical techniques. For example, when I was training 25 years ago, we were doing much longer surgeries with various contact lenses including irrigating lenses under higher infusion pressures and corneas routinely were pretty badly beaten up; it was not unusual for even healthy corneas to be susceptible to epithelial defects. The classic eye at risk for major post-surgical corneal epithelial problems, however, were in patients with diabetes mellitus with eyes having advanced diabetic retinopathy. Fortunately, nowadays with noncontact systems, shorter operating times, intraocular pressure control, and sutureless approaches for vitreoretinal surgery, we don’t see corneal epithelial problems as often as we used to. However, it still does happen, and it once again tends to be the diabetic patient but other factors that play a role in an eye being susceptible to post-surgical corneal issues include eyes with previous corneal problems, tear film abnormalities, poor lid closure, and anterior basement membrane dystrophies.

For office-based post-intravitreal injections, it is not uncommon for there to be significant corneal epithelial issues within the first 24 hours after the injection. Occasionally, it can be relatively severe and affected patients can be very uncomfortable the evening after an injection. The corneal issues after an intravitreal injection are mainly related to the prepping process, with the main culprit being an irritation effect to the cornea from the betadine. Eyes with dry eye syndrome appear to be particularly susceptible.

Dr. Kaiser: Dr. Albini, are there certain patients in whom you suspect this may be an issue, or is it really based on diabetics, longer surgeries, contact-lens-based wide field imaging, or combination cataract-vitrectomy surgeries, which seem to cause more issues than non-combination surgeries?

Dr. Albini: I would echo everything that Dr. Regillo said. I think that it is easier to maintain corneal clarity because our surgical times have gone down and we don’t use contact lens viewing systems. But there are still a couple of scenarios that I worry about. One of the most common and the one that I hate the most is the 90-year-old patient with a dropped intraocular lens (IOL) — during surgery to suture in a new intraocular lens and recover the one that fell, the endothelium can go. So, this is an endothelial problem, and losing the endothelium after a successful IOL suture surgery is very frustrating. It does not happen that often, but those surgeries can be long, and the patients tend to be older and to already have an endothelium that has been traumatized. So, that’s one situation that I’m very wary of. The other one would be the bad diabetic where you wind up with a persistent epithelial defect. Those seem to be fewer and farther between, but when they happen, it takes a long time to recover.

Dr. Kaiser: A lot of these patients are undergoing retinal surgery. So, Dr. McDonald, as a refractive surgeon, what best practices should we follow in those patients? We occasionally do need to debride the epithelium in a patient who’s had refractive surgery. Do you have any tips or tricks that you’d recommend in terms of how to avoid complications?

Dr. McDonald: First, I have to echo what Drs. Regillo and Albini said. Years ago, numerous patients were referred to me who had sloughed an entire sheet of epithelium or had it removed and now were presenting with a chronic defect. But, because of advances in retinal surgery and shorter surgery times, you don’t see much of that anymore, thank goodness. Also, we’re in a large group of surgeons now at the Ophthalmic Consultants of Long Island. We have 3 retina specialists. If they see someone, especially a diabetic but anybody with significant ocular surface disease, they send them to the cornea service for a tune-up before they go for retinal surgery, if time permits. So, they’ll get plugs, or Restasis (Allergan), or Xiidra (Shire), or whatever else they need, be it preservative-free tears or omega-3s, or night-time ointments, or microblepharoexfoliation and thermal pulsation therapy, and then they have their retinal surgery and they do well.

Instead of big disastrous complications, what I see more of are minor complications, such as when the lids are widely stretched for a retinal injection; this might cause or exacerbate pre-existing lagophthalmos. These people are older and have flaccid lids; maybe their lids are still recovering from being stretched for phaco. In some cases, they’re just starting to recover and now they need an injection. So, their lids are stretched again — with or without a speculum. Often there is significant exposure keratitis and lagophthalmos in those patients, but it’s easy to treat with frequent preservative-free tears and night-time ointments.

In patients who have had prior refractive surgery, on occasion we see late-onset of haze if they’ve had photorefractive keratectomy (PRK). This is rare now, and it’s more common if a patient had a PRK for an extreme amount of myopia or hyperopia years ago, an amount that we would not attempt to correct now. These were small, crude ablations, before the flying spot lasers; the ablations were delivered with a diaphragm that opened or closed. There were lots of steps to the ablation and this surface irregularity is not an ideal situation; anything, even years later, an epithelial insult of any kind, such as ultraviolet light, chemical, or mechanical, can set off subepithelial haze. That just doesn’t happen with people who’ve had laser vision correction more recently, with smooth ablations created by flying spot excimer lasers with trackers. But if they had PRK years ago, it can still happen.

Aside from traditional bandage contact lenses that can be used when the defect is new, it is best to debride a chronic epithelial defect, especially if you see raised white-gray edges, and insert an amniotic membrane. The amniotic membrane Prokera Slim (Bio-Tissue) is the one I use, because it is processed in such a way that the fetal healing agents such as heavy chain hyaluronic acid and Pentraxin 3 are preserved, but there are several other amniotic membrane bandages available. It’s fairly straightforward now to follow the treatment ladder and get these defects to heal. But the bottom line is, we don’t often see the more serious complications —large non-healing epithelial defects — after the big retinal procedures anymore, just the occasional minor complication due to injections.

Dr. Mah: Concerning post-LASIK patients, every once in a while, a patient will come in and, for whatever reason, the vitreoretinal surgeon didn’t know that the patient had LASIK and the flap has been lifted up as part of the epithelium removal. But if you know that the patient has had LASIK, fanning away from the hinge as the epithelium is removed is a good idea. That way, you’ll keep the flap down. Again, that’s assuming you know the patient has had LASIK and where the hinge is. The flap technically doesn’t really have any power. Technically, it’s a plano lens. If you’ve got a traumatized flap for whatever reason — it just does not look like it’s going to be viable — you can actually cut it off and technically there’s not going to be any power in the flap. I wouldn’t tell everybody to just to go ahead and cut them off, but then you have to worry about pretty bad haze, which can occur if you’ve done that.

Dr. Albini: An article was published recently by the American Society of Retinal Specialists’ (ASRS) Research and Safety in Therapeutics (ReST) Committee that suggested that whenever the epithelium is scraped it’s important to take a time out at that moment to make sure that we’re aware if the patient had LASIK or not. There have been cases where the vitreoretinal surgeon may have not been aware at all and removed the entire flap inadvertently.

Dr. Mah: I think that’s an excellent recommendation.

Dr. Kaiser: Unfortunately, it happens more often than not. During a recent ASRS film festival, one of the films showed a physician peel the LASIK flap completely off without even realizing that they were peeling the flap and not epithelium. So, all those points are excellent. Dr. McDonald, let’s assume that a patient actually had an epithelial defect from any of these indications that we’ve talked about. At what point should a vitreoretinal surgeon refer a patient to you?

Dr. McDonald: If the defect is the same size on 2 consecutive visits or bigger. If it has slowed in its progress to a significant degree — it may be getting a tiny bit smaller, like one step forward, two steps back, and the edges are gray and raised. Any defect that lasts longer than 1 week. Those are several criteria that would be helpful for a vitreoretinal surgeon to refer.

Dr. Kaiser: Dr. Mah, let’s say on postoperative day 1 you see an epithelial defect. What steps should we take at that point?

Dr. Mah: First of all, take a good history to help determine what some reasons for that might be. Is it diabetes, which is obviously going to be a much more common phenomenon, or is it something a little bit more insidious? Maybe herpes simplex or zoster is remotely in the history but may be playing a role. Maybe it’s nonsteroidals or cranial nerve issues. Make sure you haven’t missed any serious cause for an epithelial defect.

As for looking at the epithelial defect, I think that as soon as you see significant lack of clarity, the person needs a referral to make sure that there isn’t going to be permanent vision loss. As Dr. McDonald was saying, you just kind of go up the scale. You usually start with lubricant eye drops. Obviously, you’ve got an antibiotic on there. You can go up to gels and ointments as well. Then, a bandage contact lens is typically where I go.

It’s completely reasonable to refer, like Dr. McDonald said, if you’ve used ointment and lubricant eye drops and it’s not healing within a couple of days and not making that progress that you would expect. But just make sure there’s nothing in the history that you’re missing. Stopping the nonsteroidal eye drops is a pretty common cause of persistent epithelial defects or even melts. Some of the other issues, like anterior basement membrane dystrophy or a history of PRK or LASIK, are key things that I would highlight. There are some more rare causes, but these things are going to be the more common causes that you don’t want to miss.

Dr. Regillo: I have a question for our cornea colleagues. In a postoperative case, should we also make an attempt to taper off our topical steroids more quickly?

Dr. Mah: Actually, steroids do not promote melting. It’s really the nonsteroidals that promote the collagenolysis. The steroids do prevent epithelial migration, so you might want to decrease the steroids as much as you can. That would probably be prudent. But to some degree, it depends on how much steroid you need for the vitreoretinal surgery itself to make sure that you’ve controlled some of the inflammation, so you don’t get significant inflammation inside the eye.

But I do think that tapering it down a little bit would be better. There are some steroids that are in gels. For example, Lotemax (loteprednol etabonate ophthalmic suspension; Bausch + Lomb) is in a gel. If you’re contemplating something that you’re going to decrease down to, that might help protect it. It can increase the contact time, but reduce some of the dosing of the eyedrop. Or, you may want to jump up to something that’s more potent, such as Durezol (difluprednate ophthalmic emulsion; Novartis), but decrease the dosing. You can go down to once or twice a day instead of the 4 times a day or every 2 hours with prednisolone acetate, which actually has a lot of preservative in it if it’s the generic.

Dr. Kaiser: A final question for the cornea specialists: do you have tips for preventing epithelial issues after using betadine swabs for intravitreal injections? What’s the best way to deal with corneal discomfort after intravitreal injections?

Dr. McDonald: You can’t go wrong with preservative-free tears at least every 2 hours while awake, for 1 week afterward, with a bland ointment at night. That helps because almost all patients, unless they’re very young, are going to get traumatized lids again. They’re going to get a little lagophthalmos and exposure keratitis.

Dr. Mah: Another thing that you want to do, at least in the clinic, is try to minimize the time that the proparacaine and betadine have been sitting on the eye. Maybe have the eye closed while the patient is waiting for the injection. Immediately afterward — I completely agree with Dr. McDonald — using some preservative-free lubricant is an excellent idea.

Dr. Kaiser: Any last comments from our retina colleagues? I certainly have learned a lot today.

Dr. Albini: Anecdotally, as I’ve been switching over more of my patients to not using a lid speculum for routine injections, the amount of corneal abrasions that we’re seeing afterward has gone down. To me, that was always one of the most dreaded things that would happen more commonly after an injection. I don’t know if anyone has looked into that, but I think that a good number of those abrasions were caused by the lid speculum. Moving away from the lid speculum will be helpful in reducing abrasions in this setting.

Dr. Regillo: I agree. Our group stopped routinely using the lid speculum a few years ago and we’ve definitely seen fewer post-injection problems. We also moved away from lidocaine gels which appeared to prevent the patient from fully closing their lids for a while. It turned out to be that less was more — the less manipulation, the less we did, streamline the prep — the better things got. A drop of proparacaine followed by a drop of betadine, that was just mentioned. Don’t wait a long time and keep the eyelid closed. We sometimes even put some artificial tears in before the prep if the patient has a dry eye or has had corneal problems to lubricate it ahead of the prep. All these measures seem to be beneficial in helping to minimize the problem.

Dr. Kaiser: We’ve covered a lot of bases. Retinal Physician has begun a series of multidisciplinary roundtables, and I thank both Drs. McDonald and Mah for joining us, because we really learn a lot from other subspecialties. I appreciate everyone for giving us their time. RP