Lecture: Penetrating Keratoplasty in Infectious Keratitis

Infectious keratitis is a common corneal condition worldwide. Causes vary by region but can often lead to corneal opacity and permanent vision loss. In severe or non-responding cases it may also lead to perforation or progression to endophthalmitis. In certain cases, therapeutic keratoplasty is needed to control the disease or treat a perforation. In cases that can be controlled medically, residual corneal scarring may require keratoplasty for optical reasons. This webinar discusses the indications, techniques and outcomes of therapeutic keratoplasty in infectious keratitis. It also focuses on timing of surgery and special considerations the corneal surgeon should take into account when performing these procedures.

Lecturer: Dr. Soledad Cortina, MD, Associate Professor of Ophthalmology and Visual Sciences, University of Illinois Eye & Ear Infirmary, Chicago, USA

Transcript

DR CORTINA: Okay. I am Dr. Cortina from the University of Illinois, Chicago, and today we’re going to talk about corneal grafting in infectious diseases. I have no financial disclosures to make on this topic. So why are we talking about grafting in infectious diseases? Well, if you think about it, and look at the epidemiology, about 20% of corneal transplants done worldwide are actually for infectious keratitis. And of course, the indication for corneal transplantation varies according to the regions in the world. So 24% of all countries in the world actually the primary indication for transplantation is infectious keratitis. There are more potent antibiotic agents right now available, so the incidence of transplantation to control bacterial infections is decreasing. However, for fungal ulcers, surgical treatment is still a large part of the management of these conditions. So how does corneal transplantation work in infectious keratitis? Well, when we’re talking about a bacterial infection, more often than not we might be able to completely remove the infected tissue by performing corneal transplant. However, in cases of fungal or acanthamoeba infections, removing the entire infected area may not always be possible, but debulking the infection might allow antimicrobials to eradicate the residual infection in the host. Let me see how I advance. Oh, there we go. So who is a candidate for a corneal transplantation? How do we select patients, or how do we indicate a therapeutic keratoplasty? Well, number one is preservation of the globe. So if we have a patient that has a perforation, then that’s a no-brainer. That needs to be closed. He needs a transplant. Or maybe he may have an impending perforation, and we want to get in there before things get more tricky. Because it’s always tricky to do a transplantation in a patient that’s already perforated. When we’re starting to suspect extracorneal extensions, that’s an indication that the infection is really not under control, and we need to do something to try to stop it. We’re thinking that there is some intracameral extension or endophthalmitis. That might be an indication. Or impending scleral extension. Because once it gets into the sclera, that’s really difficult to remove surgically. That’s in the acute phase, when we’re trying to control the infection. But the other indication would be an optical graft for visual rehabilitation after the infection is controlled. And as we’ll see later on, when we discuss all the evidence and the published studies, of course outcomes of optical grafts are much better than those done with a hot eye. So successful medical therapy is always better. So we’re not gonna rush into corneal transplantation. And actually, medical therapy for all forms of keratitis, even for fungal, acanthamoeba, bacterial, viral, have improved dramatically. Medical cure rates have risen. There are new options and tools for management and diagnosis of these conditions. So I think that to try medical therapy and to try alternative medical therapy before we move into surgical procedures is always best. This is the first question on our poll, and the question is: What are the most common causative organisms requiring therapeutic keratoplasty? All right. So this is a little bit of a tricky question. Most of you answered fungal keratitis. And then second to that was bacterial keratitis. So actually, if we look at the total percentage or the total number of grafts being performed, the most common indication would be bacterial keratitis, with 31%, and that’s actually because bacterial keratitis is much more common than the other forms of keratitis. But if we look at the rate in all bacterial keratitis and all fungal keratitis, so obviously more fungal keratitis might require transplantation in percentage. So we have 31% for bacterial keratitis. Mainly being pseudomonas, because of the virulence and coag-negative staph. Mixed ulcers are 6.9%, viral are 5.3, and acanthamoeba is 1.6. This just relates with the rarity of the acanthamoebal infection. Non-invasive surgical treatments are to be sought. So conjunctival flap, although that is a surgical treatment, it’s not a penetrating treatment, and that can be considered in some cases. Tarsorrhaphies, collagen crosslinking, there are even studies showing argon laser, trying to break up the biofilm of the ulcers to let the antibiotics or the antimicrobials penetrate better. And cyanoacrylate glue in those that may progress to perforation. But always remember that bacterial infections may still progress under glue. And in a study performed by Sharma et al. in India, they found that out of 506 eyes close to 90% — they had a close to 90% success in restoration of tectonic integrity when ulcers perforated. So let’s talk a little bit about keratoplasty in bacterial infections. As I said before, pseudomonas aeruginosa is probably the organism most likely associated with the need of a keratoplasty. It may produce collagenase, with rapid corneal thinning, and can progress to perforation in 24 to 48 hours. So perhaps it perforates. It doesn’t necessarily mean that we cannot get this infection under control medically. By eradicating pseudomonas, because we do have antibiotics that can treat it well. It’s just that the cornea thins and melts so quickly that sometimes we just don’t get in there in time. And then the other one is crystalline infectious keratitis that can be associated with candida or alpha hemolytic strep, and in fact, 50% of these types of keratitis fail medical treatment. The reason is it is so deep in the stroma, it is hard to get to it. In combination with medical therapy, the cure rate of infection for bacterial infections reaches 90% to 100%, which is very good. Hard to say what graft to do, which technique. Should we do a lamellar or should we do a penetrating graft? So in this study, published in Ophthalmology in 2009, we saw 126 eyes undergoing therapeutic keratoplasty. There were 26 therapeutic it TDALKs, versus 100TPKPs. We saw the recurrence rate in the TDALK group was higher compared to 12% in the TPKP group. And we saw that 9 out of the 12 therapeutic TPKPs required evisceration. The infection was that bad. Of the TDALKs, we saw they had better graft survival and vision overall. So slightly higher recurrence, but better overall outcomes with the TDALKs. So if the infection is not that deep, perhaps this is the way to go. Fungal keratitis. What are the things to consider when we’re thinking about the need of therapeutic keratoplasty? Well, perforation is common in these ulcers, in up to 50% of cases. We’re gonna think about penetrating keratoplasty or keratoplasty when there’s lack of response to medical treatment. We do for sure want to do very intensive antifungal therapy, prior to penetrating keratoplasty, because these will help stabilize the cornea, and will improve the long-term prognosis. Remember, fungal elements can actually penetrate through Descemet’s membrane. So I think that in fungal keratitis, one would opt more towards therapeutic keratoplasty than a lamellar keratoplasty, because there might be some higher recurrence in lamellar grafts. Pollack showed that PKP was actually far superior than lamellar keratoplasty. There is another large Chinese study that looked at 604 therapeutic PKs, or keratoplasties for all causative organisms. They obtained 95% globe preservation, but they had a recurrence of 3.5 percent in penetrating grafts and high recurrence, 7.3%, almost double, for lamellar grafts. So think that therapeutic keratoplasty is not always curative. So the organism may come back. The infection may recur. In fact, fungal keratitis may have a recurrence rate in some studies — the one that we looked at before had a lower recurrence rate, but the one from Singapore showed a 47% or higher recurrence rate. These looked at 92 consecutive grafts. 15 of the eyes experienced recurrences. 11 out of those 15 were fungal infections. And most went on to evisceration. So if the fungus recurs in the graft, that’s usually not a good sign. Risk factors of recurrence were scleral extension and perforation. So this is our second poll question. Which is not a predictor of need of therapeutic keratoplasty in severe fungal keratitis? Great, so I think most of you chose visual acuity, and that’s absolutely right. If we go to the next slide, this is an important study that was published in JAMA in 2017. It was also coming out from the Singapore registry. And what it basically showed — they defined high risk features at baseline as an infiltrate that was bigger than 6.6 millimeters, that it was involving the posterior 1/3 of the stroma, and that had a hypopyon. So those were considered high risk features for perforation or for the need of therapeutic keratoplasty. When they did the multivariate analysis on patients with fungal keratitis, what they found was that the depth of the infiltrate, the size of the infiltrate, and the presence of a hypopyon, all of that was statistically significantly associated with the need of a therapeutic keratoplasty or a progression to perforation. However, visual acuity and the other factors that they looked at — agricultural occupation — they did not find to be related. So all of these patients came out from the Mycotic Ulcer Treatment Trial II. So the high risk patients had ulcers of more than 6.63 millimeters, and the low risk were the other patients in the same trial that did not have these characteristics. So it was infiltrate depth, infiltrate size, and hypopyon. Let’s continue to talk about therapeutic grafting in fungal keratitis. What about postoperative management? Someone from the audience — one of the questions was: When do we use corticosteroids? And I would say caution when we’re talking about fungal keratitis. It is better to withhold, because it is a risk factor for recurrence in the graft. There is some small series looking at cyclosporine A instead, and they have shown safety, but again, this is a small number of patients, and it appears that cyclosporine might be suppressive on fungal growth, so perhaps this could be an alternative to corticosteroids, to try to preserve the graft, and at the same time not impair the treatment of residual infection. This is just a case of mine. It’s a 58-year-old male with very severe facial thermal burn. He had a Boston type I keratoprosthesis in the right eye about 7 years ago, and he has significant exposure from cicatricial ectropion. His vision is about 20/200, and this is his baseline. He comes back one day looking like this, and you can see all this pigmented stuff on the cornea. And this is a zoomed photograph. You can see how it’s surrounded the cornea, and he has a contact lens over. So basically once we needed to do a therapeutic keratoplasty, because the infection had gone — basically eaten through the cornea, and you can see at high power, this is a PAS stained section showing a mass of fungal organisms, and at medium power, you can see the perforation, and you can see the fungal organisms. So we replace it with another keratoplasty, and the culture showed exophiala phaeomuriformis. We treated with oral voriconazole and topical amphotericin B, and six months postop he had no recurrences and the vision was improved. So things to think about when dealing with infection and a prosthetic device is that sometimes clearing the organisms around a prosthetic device might be more difficult, there might be biofilm forming over the surface of these devices, it may be difficult for antibiotics and antifungals to break, so consider therapeutic keratoplasty earlier in the course of the disease than you probably would in a patient that doesn’t have a prosthetic device. And perhaps in this case we will replace it with another Kpro, but sometimes when you’re not sure you’re able to control the infection, it might be better to replace it with a regular cornea, knowing that it will fail, and go back to the Kpro later, once you’re 100% sure the infection is controlled. You have to consider, because of the surface, the plastic that the keratoprosthesis occupies on the surface of the cornea, the penetration of the antimicrobials might not be optimal. What about acanthamoeba keratitis? Well, the efficacy of the medical treatment for acanthamoeba keratitis was looked at in a study published in Cornea in 2014. They looked at 116 patients, and they saw that 66% were successfully treated medically, but 33% actually required keratoplasty for the treatment. Ficker et al. showed that there is poor graft survival in patients with acanthamoeba keratitis. 50% had recurrence in the graft. Other series had somewhat better outcomes than shown in this study. The most common complication after therapeutic keratoplasty is glaucoma, mydriatic pupil, and graft failure, about 50%, like we said before. So optical keratoplasty is always better. If we can wait, if we can control the infection medically, these will always be the better outcome. So this is another study in acanthamoeba keratitis. They looked at 22 eyes, 20% had a therapeutic PKP, two had a lamellar keratoplasty, versus 9 eyes with an optical penetrating keratoplasty. We saw that the recurrences were 41% in the therapeutic keratoplasty, but interestingly, it was 22% of recurrence in the optical keratoplasty. So when we thought that perhaps the infection was controlled, was eradicated, did the optical keratoplasty, and still in acanthamoeba, we saw recurrences up to 22%. So repeat grafts, 55% of therapeutic keratoplasties needed to be repeated, and only 11% of the optical keratoplasties needed to be repeated. So we look at graft survival at one year. Always the optical is better. When we go out to 10 years, again, about double the success rate for the optical keratoplasty, compared to the therapeutic one. And of course, this is to be expected. Some other studies on penetrating keratoplasty in acanthamoeba. There’s a study out of Dallas that looked at 13 eyes. They saw that the best corrected visual acuity was 20/15 to 20/40 in 13 eyes. Excellent outcomes. One had light perception secondary to glaucoma, and all 12 remained clear for a median of 23 months. Another study out of Brazil saw that 55% of grafts were clear at one year, they had two recurrences, and they had a very high rate of glaucoma, which they identified as the primary risk factor for graft failure. So this is another case I wanted to share with you. This is a 68-year-old man with a history of keratoconus, and rigid gas permeable contact lens wear. When we see him at 2 months, he reports discomfort and non-healing epithelial defect, he had been treated for HSV keratitis, neurotrophic ulcer, including amniotic membrane transplantation, and I apologize for the blurriness of the photograph, because this was taken with the patient’s phone. He looked like this when he came in. It was kind of a diffuse infiltrate of the whole cornea. This was a confocal microscopy, and you can see it is actually full of amoeba cysts. So he was started on brolene, propamidine, and PHMB, and we also started him on oral voriconazole. You can see how the limbus looks a little bit heaped up, and we had a question on whether there was actually limbal extension of the infection. Unfortunately he develops very difficult to control glaucoma. Every complication that we talked about before you’re gonna see in this patient. Very difficult to control glaucoma, mydriatic pupil, rapidly progressive dense cataract. And these also might be side effects from the topical agents that we’re using. A shallow peripheral chamber with development of anterior peripheral synechiae, the epithelial defect conditions to enlarge, the cornea thins, he has an impending perforation, we try to do amniotic packing, but he returns with pain, a flat chamber, a high IOP, and a perforation. We take him to the operating room, we performed a 10.5 millimeter PKP, almost limbus to limbus, we do a peripheral iridotomy, take out the cataract, we place a sulcus IOL, put an Ahmed valve, all at the same time. This is the corneal button that we removed at the time. And you can see the load of organisms in this here is massive. And actually, the amoeba went all the way up to the margins of the trephination, so we know we have residual disease. He returns two days later with an IOP of 50, he has pain, the IOL is touching the graft, very flat chamber, so we think that he has aqueous misdirection. We take him to the operating room, do a pars plana vitrectomy with a posterior capsulotomy, and another peripheral iridotomy, and he does well after that, but we had to do a one-month course of miltefosine. This is a new drug. I think there are some studies coming out, which this patient is actually a part of, looking at refractive cases of acanthamoeba keratitis using this drug. He used this for two months, I’m sorry, one month, plus a two-month course of topical PHMB, and the graft has been clear so far, with no recurrences. So this was an extreme case, very advanced case of acanthamoeba keratitis, with a good outcome. Now, this is only 6 months after we did this graft. What will be his one year, five year, ten year survival? Probably not great. Optical keratoplasty for acanthamoeba keratitis is another case. He’s a truck driver. He presents with classic acanthamoeba signs. He responds well to medical treatment. We do an optical keratoplasty. And despite the fact that he has some neovascularization around the graft, he did very well. No rejection, no recurrence for over five years. Therapeutic keratoplasty in herpetic keratitis. So HSV is most commonly performed — keratoplasty in HSV is most commonly performed for optical purposes. I would say the vast majority would be for optical purposes. But there are some cases of active stromal disease that have significant ulceration or perforation, and maybe removing the antigenic material causing repeated immunologic episodes might be worthwhile. It does require oral acyclovir therapy and topical steroids. Wang et al. published 22 cases of lamellar keratoplasty in active disease, and he had one graft rejection and 4 recurrences. Herpes zoster, the prognosis as you know is generally worse than that of herpes simplex virus. There’s more severe anesthesia, and it usually requires adjunctive tarsorrhaphy and amniotic membrane transplantation to get the eye to reepithelialize. This is another case of mine, a Kpro that developed very severe HSV keratitis, And the reason I realized this was not just a melt in the cornea but HSV is that you can see that the infiltrate doesn’t just involve the donor cornea, but also the host cornea. She actually had severe vitritis related to HSV keratitis. She responded very well to oral acyclovir, and she went back to baseline. I think after almost thinking we were gonna have to do a therapeutic keratoplasty in her, medical therapy was eventually successful, and she ended up with a very thin cornea that held, and we didn’t have to replace it. What about preoperative planning? Make sure you record the size and the depth of the corneal infiltration, the presence of limbal involvement, to help you decide how you’re gonna place that keratoplasty, whether there’s corneal perforation or not, whether there’s associated endophthalmitis or glaucoma, and you need to do something about that during the surgery. Try to do a dilated fundus examination, if there’s no view. Make sure you do an ultrasonography to rule out the presence of vitritis. Systemic and topical antimicrobial treatment. If the causative diagnosis is not established, you want to introduce broad spectrum antibiotic or combination therapy. Hopefully the pathology tissue that you’ll recover from the keratoplasty, if you send it for culture, and you send it for pathology, perhaps PCR, you might be able to now to identify the organism and direct your treatment a little bit more accurately. Perioperative treatment. Intravenous mannitol to decrease intraoperative complications due to pressure abnormalities might be helpful. Injections of anti-VEGF agents might be helpful. What anesthesia to use? General anesthesia is preferred on a perforated globe. Peribulbar injection is a viable option in cases without perforation. It is really important to try to have a soft eye that is not moving, and where the anesthesia is well achieved. However, there might be patients that are not candidates for general anesthesia, and Riddle et al. published topical anesthesia in eight cooperative patients with corneal ulcers and associated systemic comorbidities. So while I don’t advise this approach, it can be done in extreme cases. What about the donor material? What kind of tissue should we order? Or should we take? Well, you want good quality donor tissue. A healthy donor endothelium will help you maintain a clear graft, despite the associated inflammation, and the intraocular pressure. So it will give you better chances of success. And an intact epithelium — I think this is important, because it minimizes the risk of graft reinfection. So it’s not that often that we get corneas with intact epithelium. But if you have one, I think this will be better. What if you don’t have readily available fresh tissue? We can use non-optical grade cryopreserved tissue. These are corneas that are devoid of live cells. The chances of an allograft rejection are minimal, because the antigens are minimized by the absence of cells, so there’s no need for steroid use. Once the infection is resolved, then you can go in and do a smaller diameter optical keratoplasty, and rehabilitate the patient visually. Though, again, corneal tissue is scarce in many parts of the world. There’s a limited supply of transplantable corneas for DALK or PKP. So there was a study published in Cornea in 2006, that looked at the use of preserved tissue. They used 22 grafts, preserved, 16 of the 22 grafts were for fungal infections. Complications that they found were glaucoma, wound leakage, recurrence, enucleation or evisceration — 40%. A high number. And the final visual acuity was 1/60. What if we do cryopreserved versus optical grade tissue? 10 eyes received cryopreserved tissue followed by optical grade tissue, and 22 eyes received optical grade tissue alone. 72.9% of the cryopreserved were successful and 53.8 of the primary optical were successful for one-year survival. So basically we’re using this cryopreserved tissue when we have this high inflammation. We’re giving all these drops that can be somewhat toxic, and then once the infection is resolved, shortly after, we go to fresh cornea, and then that fresh cornea will have much higher rate of success than if we did it from the get-go. This is what the study — the bottom line of this study, which was done out of the Singapore group. What about the surgical technique? And some intraop considerations. So think about support. Maybe a Flieringa ring might be important. Avoid pressure on the globe in cases of perforation. And for this reason, think about what kind of speculum you’re gonna use. Sometimes they can put pressure on the globe. Maybe you want to use a Schott or Jaffe or different kind of speculum to retract the eyelid. Perhaps even sutures. Try to remove all the infected tissue, leaving about a 1 millimeter rim of healthy tissue if possible. Even if you have to do an eccentric or large graft, remember that you’re getting in there on a hot eye to control the infection. Careful trephination. Glue sometimes can be useful if you have a perforation, to try to pressurize the globe and obtain a better trephination, but you can also do handheld or free handed trephines sometimes. Take cultures. Especially if you don’t know what you’re treating. But even if you do, it never hurts to take cultures. Send tissue for pathology and microbiology, and consider, like we talked before, intrastromal, subconjunctival, or intravitreal injections, as needed. Remember that less is more. If you don’t see the crystalline lens affected, leave it. It will be a barrier between the anterior and posterior chamber. Irrigate the chamber with BSS, inspect the iris for infectious foci, address risk factors for perforation, and consider peripheral iridotomy. Why? Because this is a hot eye. And it is possible that you’re gonna have a lot of inflammation after these transplants, so you may have syneched pupil that may cause secondary angle closure, and if you have peripheral iridotomy in place, this is helpful, when you have a bad view on a hazy graft and visualization on the anterior chamber may be difficult. This is a case of enlarged diameter corneal-scleral grafts, one of those hail Marys when you’re trying to control infection that is very advanced not just in the cornea but also the sclera. In this case, we need almost a complete anterior segment or corneal-scleral transplant. In an advanced case of acanthamoebal keratitis. She’s 2 years out with 20/30 visual acuity, doing very well. This is a 45-year-old truck driver, corneal fusarium ulcer, not responding to medical treatment, oral fluconazole and topical natamycin with epithelial plaque. So what is the next step in management? And again, don’t feel bad about the answers of these questions. Sometimes there’s not only one correct answer. It’s just a question to provoke thought and discussion. All right, so it’s kind of close, but I think most of you chose therapeutic keratoplasty. Collagen crosslinking — the results are kind of mixed as to the success for that. And intrastromal injection of amphotericin B can also be an option, but if we already see that it’s going into the anterior chamber, perhaps something more aggressive needs to be done. And this is what we decided to do. A therapeutic keratoplasty. But as you can see, the aspergillus recurred in the periphery of the graft. On the area of the limbus and part of the sclera. So what to do now? And what we did was, again, another hail Mary, basically to try to preserve the eye, and we did this almost like a keratolimbal transplant or allograft, where we took part of the sclera and part of the cornea, and we moved all of that infected tissue, and we placed that over there, and then we advanced the conjunctiva over that area. And after a few months of continuous medical treatment, he eventually healed. Yeah, he doesn’t see well out of this eye, but he was able to keep the eye. We saved the eye. And perhaps later on we can do some form of visual rehabilitation. With this densely vascularized cornea and failed graft, failed limbus, he’s not a candidate for a traditional keratoplasty, but perhaps a keratoprosthesis might be an option for him. What is the prognosis? Well, in general, when we do therapeutic keratoplasty for infectious diseases, the long-term graft survival is poor. Singapore corneal transplant study looked at 901 eyes. And 8.1 of all of those eyes were classified as therapeutic. And they showed that when we look at optical keratoplasty, the success rate, again, was double that of the therapeutic keratoplasty at every single point, including 1, 3, 5, and 10 years. So what are the most significant risk factors for graft failure after therapeutic keratoplasty in infectious keratitis? So, again, one could say that all of these are causes, are risk factors for graft failure. But what do you think are the most significant? And the correct answer for this question is actually based on a study. So it doesn’t mean it is the absolute truth. So most of you chose presence of corneal inflammation, but all of you thought of some of them. So we’re looking at the Singapore study. When they looked at the risk factors for graft failure, they found that corneal perforation, number one, and the presence of ocular inflammation are the most significant risk factors for graft failure. And we all know that corneal neovascularization means a high risk graft and that glaucoma and corneal grafts don’t get along very well, so by all means, they are all risk factors, as you can see by the Kaplan-Meier curves, where the presence of glaucoma has less survival, the presence of perforation has less survival, the vascularization and the inflammation, but of all of them, those two were the most associated. So key points of everything that we’ve talked about today: Most common indication is still bacterial infection, just because it’s the most common infection. But the rate of fungal and acanthamoeba keratitis requiring grafting is approximately 50%. Success rates of surgical therapies are best for bacterial corneal infections and worst for fungal infections. Unresponsive corneal infections, impending perforations, and perforations are the usual indications for surgical intervention. Tissue is scarce and cryopreserved tissue is a good option. Long-term graft survival is poor, and better when performed for optical indications. So I thank you for listening. My email is here if you ever need to reach out for anything, but we can answer questions you might have.

DR CORTINA: Okay. So the first question says: What is the cause of keratitis? I’m not exactly sure I understand this question. Perhaps whoever wrote it could clarify. The second question: In which type of infectious keratitis penetrating keratoplasty is more successful. And the answer is in bacterial keratitis. How under glue bacterial keratitis may increase? So we use the glue to perhaps prevent a perforation or to treat a small perforation. But it doesn’t necessarily treat the infection. And under the glue where the antibiotic might not be penetrating, the infection may progress. That’s what I meant when I mentioned that. What methods of suturing, interrupted or running? I would say always interrupted. Because there is inflammation. There is gonna be wound healing, and the sutures may loosen much sooner than in a regular graft. So having interrupted sutures will allow you to selectively remove some of them, versus just the running suture. Perhaps where you’re dealing with vascularization and things like that, I think an interrupted suture may allow you to revise the wound if you have to, in some areas. I think that is the better technique for these circumstances. For corneal-scleral grafts, is the scleral portion partial thickness? Yes. So if you can thin it out, that’s usually better. You can do a sclerectomy on the host, remove partial sclera, and then try to thin out the donor sclera, and then you can suture them together. And I think that works best. Almost like you’re shelving them.

DR CORTINA: Okay. Do prophylactic peripheral iridotomy sometimes increase infection? No, I don’t think so. What I mentioned by this is: Whenever I’m doing a penetrating keratoplasty on an eye that’s inflamed, that’s actively infected, I do peripheral iridotomy, because I will expect there will be inflammation, perhaps fibrin, in the anterior chamber, in the immediate postoperative period, and having this peripheral iridotomy decreases the chances of having angle closure and pupillary block after surgery. My question: How long we should wait before deciding for therapeutic keratoplasty? I think this is a great question, because timing is everything. You know, you don’t want to do it too soon. You want to allow the antimicrobials to work, because like we said, if we can control an infection medically, that’s always better. But if you wait too long, you may have lost that window of opportunity where the keratoplasty could have taken care of the situation by removing most or all of the infection, and then if you let it advance to sclera or go into endophthalmitis, and perhaps it’s too late and the success won’t be as good. So timing is everything. So if you have an infection where you tried the antimicrobials, it didn’t work, you switched to whatever alternatives there are and it’s still not working and you’re seeing it progressing or you’re starting to see the potential for extracorneal extension, then by all means, I think that’s the time to go ahead and do the keratoplasty. The next question is: What if in the few days postkeratoplasty you have a suture infiltrate? Would you remove the suture with the risk of leaking? The answer is yes. I think I would remove the suture, if I have to replace the suture, then you can replace the suture. That’s why I think interrupted sutures are much better. Because oftentimes you can remove just one suture, and not have a leak. Next question: Where do you obtain PHMB and brolene from? Is it compounded? Yes, our pharmacy compounds the PHMB and the brolene we obtain from Europe. Next question: How does the outcome of collagen crosslinking compare to therapeutic PK in bacterial keratitis? I don’t think I have an evidence-based answer to this question. I don’t think there are studies comparing penetrating keratoplasty to crosslinking directly. I do think that studies done for crosslinking on advanced bacterial keratitis are a smaller number of patients. The other thing is: I think that collagen crosslinking, perhaps in the future, might have a bigger impact in early ulcers than it does in advanced ulcers, where penetrating keratoplasty is always for more advanced ulcers. The exact role of collagen crosslinking in infectious keratitis — I think that’s yet to be determined. We don’t know exactly what the role is. The group out from Sweden that’s advocating the use of collagen crosslinking as the initial treatment for ulcers, for example, for patients that have poor access to medical therapy, or to seeing a doctor, then they can come in, perhaps they traveled six hours to see the doctor, they have a corneal ulcer, so could we do collagen crosslinking as the sole treatment for this small ulcer, and take care of that without the patient having to use antibiotics or come back for frequent follow-up? This is a hypothesis that they’re working on. There are some studies like I talked about, where they used this for advanced non-responsive ulcers, and in some of them it works somewhat, and in some of them, it doesn’t. When do you suspect that hypopyon is infected rather than a sterile one? This is a good question. And sometimes it is difficult to determine whether the infection has progressed into the anterior chamber or not. I think that for the most part, if we’re talking about pseudomonas bacterial ulcer, usually the hypopyon tends to be sterile. When we’re looking at a fungal ulcer, it’s more commonly that you’ll see these endothelial flaps that denote intracameral infection or extension of the infection. So I think the organism is usually more suspicious when we’re talking about parasites or fungal. That the hypopyon is actually infectious, rather than sterile. But ultimately, you never know for sure, unless you sample the anterior chamber. Next question: Is it possible to try amniotic membrane transplant first for impending perforation, before considering therapeutic keratoplasty? And I think the answer is yes. As you saw in one of the advanced acanthamoeba patients, that’s what we tried. The amniotic membrane packing, to avoid the perforation and see if we get the epithelium to heal. That didn’t work and ultimately we had to go to therapeutic keratoplasty. So as long as the patient understands that there might be another surgery soon, if this doesn’t work, it can be tried.

DR CORTINA: All right. So some of the questions we went through that were sent ahead of the webinar. We’re just gonna go through now… The first one is: Advice on timing of PKP, when culture positive infection with perforation? I think once it’s perforated, my gut would be just to go right away. Once the patient is perforated, then I would want to do the PKP right away. The same day, if possible. Can this be done for a patient with a hypopyon ulcer? The answer is yes. I would do a penetrating keratoplasty, wash the anterior chamber, remove the hypopyon, wash with BSS, and perhaps inject intracameral antibiotics if needed. So how to prevent the infectious keratitis? That’s a difficult one. I think this one depends on the epidemiology of each country, and what the most common causative organisms are, and what kind of public health measures can be taken to prevent infectious keratitis. In some cases, for example, here in the US, there were big breakouts of infection with fusarium, for example, secondary to contact lens solutions. So having a stricter organism to approve these types of solutions for contact lenses might be a good idea, but I think it’s different for each country. The next question is: In case the graft gets reinfected with the same organism, when would you decide to take up for re-PK? So again, the graft may get reinfected, and this infection we might be able to manage medically, but I think I would have a low threshold to go back and repeat the graft if the infection continues to be not controlled. Is there any contraindication in infectious keratitis? I assume that this is: Is there any contraindication to keratoplasty in infectious keratitis. And I would say no. I mean, you have an infection that is not controlled. So you want to try to control it, and the alternative would be that the patient loses the eye, because the infection continues to progress. So I would say no. If we’re talking about an optical keratoplasty, the traditional teaching is that a cornea that has poor corneal sensation is a poor candidate for a penetrating keratoplasty. A cornea that has a high degree of vascularization, like that patient that I showed you after the fusarium, those are very high risk for penetrating keratoplasty, and perhaps it is actually contraindicated, because we know they will do very poorly, and perhaps we have to look at other options to rehabilitate them, like an artificial cornea. Is there any update treatment of infectious keratitis? I think there is plenty, more than I can cover just today in this webinar. But there are new medications. We talked a little bit about miltefosine for refractive acanthamoeba cases. There is the mycotic ulcer trial that looked at the oral voriconazole, showing that it is not necessarily effective for fungal keratitis, and perhaps doesn’t need to be used. The most effective for aspergillus is actually natamycin, so all of these you can find in some of the mycotic ulcer trials. I think that’s a good review for the treatment of fungal keratitis. There’s also another study, again, multicenter, which looked at the use of steroids in bacterial ulcers. I think that’s also something nice to look, where they found that it is — the use of steroids is actually contraindicated in nocardia, and it might be helpful in large central ulcers, but it has no difference in the final outcome in the rest of the bacterial ulcers. New treatments for keratoplasty rejection? The traditional treatment is steroids. It’s what’s going to act faster. You can try to use a more potent steroid like here in the United States, we use Durezol, and we can use that, and then alternatively you can use cyclosporine to prevent rejection. Not necessarily to treat it. Topical cyclosporine, 1%. And we have the use for rejections that are refractive to topical steroids, which have used tacrolimus. 0.3%, topically, in drugs compounded by our pharmacy, and we’ve had some success with that. So now for all of my high risk transplants, I may put them on tacrolimus, together with a steroid. It’s also helpful to have something like cyclosporine or tacrolimus on board, even if topically, because it may help to reduce the dose of steroids we need to use and reduce the incidence of ocular hypertension and secondary glaucoma. Next question: Should anterior OCT be performed previously on all penetrating keratoplasties undergoing patient to minimize higher order aberrations effect after surgery? I think when we’re talking about penetrating keratoplasty for infectious keratitis, not necessarily. I don’t think I performed OCT. Unless I want to see something on the anterior segment to help me with surgery, but not for higher order aberrations. What does pannus signify in a healing stage? That there is a lot of wound healing going on. We don’t want vessels on our corneas, so when we’re starting to see neovascularization, perhaps we can consider anti-VEGF therapy, either topical drops or subconjunctival injections, or increase the dose of steroids. What is the modern treatment of fungal keratitis? I’m not sure how to answer that. I think that perhaps performing PCR for early diagnosis of the causative agent is the way to go, and then using the newer antifungal agents like the azoles, and caspofungin is another antifungal we’ve been using, and just early diagnosis — I think that’s key. What is the most common cause of keratitis? I think that’s bacterial keratitis, all around the world. When to decide to do the therapeutic keratoplasty? We went over this. It is when the ulcer is perforated or has imminent perforation, where there is extracorneal extension, or when the infection is progressing and not responding to medical treatment. When do you start topical steroids postkeratoplasty for fungal keratitis? Any role of systemic steroids? I think caution, caution, caution when you’re using topical steroids postfungal keratitis. I would prefer that… That the graft fails and we have to repeat it, if we have to, because it’s rejected. But I don’t want the infection to go out of control because of the use of steroids, so I think I would be very judicious in the use of steroids after fungal keratitis. And the last one I think we already talked about. When should we go for therapeutic keratoplasty in cases of fungal keratitis? And I think it’s: Think about it early. We already know that 50% will need penetrating keratoplasty. So if you feel this ulcer is not going anywhere, and it’s enlarging, get in sooner, before the infection gets too large, and then you’re not able to remove it with the keratoplasty. I think that’s all of the questions. I see one more question here in the Q and A, and the attendee says: In limbus to limbus therapeutic keratoplasty, is there a risk of epithelial downgrowth? If yes, how to prevent it? I think that… I mean, there could be a risk. I haven’t particularly encountered this complication very commonly after large grafts. But I think having a good wound, good opposed wound, with no leakage, with no gaping or anything like that, is key. Sometimes this is difficult in an active infection, because we may be suturing on tissue that is not that healthy. But I think trying to have a watertight wound is important.

Download Slides

PDF

January 11, 2019

Last Updated: October 31, 2022

Leave a Comment