Lecture: Medical and Surgical Management of Ocular Chemical Injury

Chemical injury to the eye can result in devastating vision loss. Prompt treatment in the acute phase to reduce ocular surface damage is important but visual rehabilitation in severe cases can be difficult. This webinar will briefly discuss assessment and treatment of the acute chemical injury and focus more in depth on medical and surgical techniques for the long-term management and visual rehabilitation of this challenging condition.

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


DR CORTINA: Okay. Good morning and good evening to everyone. My name is Soledad Cortina, and I’m an associate professor at the University of Illinois at Chicago today, and in this webinar we’re gonna talk about medical and surgical management of chemical injuries. I have no financial disclosures that are related to this topic. So a little bit of background. Why is this important? Well, actually, it’s 20% to 30% of all ocular injuries, are due to chemical injuries. It is defined as a corrosive substance that is accidentally introduced into the eye or the periocular tissues. And we say that in ophthalmology, there are not many true emergencies. Well, this is one. And it does require immediate and intensive intervention, to try to minimize complications that can actually be devastating, and can give the patient lifelong disability. It is most prevalent among young males, but children are also at risk. In fact, there’s been just a nationwide campaign launched in India to try to change the packaging for these lime salts that are used that typically children go and buy these salts at the store, and they’re playing with it, the package opens, it gets into their eyes, and has devastating consequences. So ocular chemical burns — it’s a variety of substances with different pHs, as you can see. And common offending agents can be at home, like cleaning solutions, drain cleaners, oven cleaners, bleach, battery explosions, industrial, lye in drain and in cement, and acids like hydrochloric acid, for example. The severity of a chemical burn is determined by different factors. One is the pH of the substance that goes into the eye. And the specific reactivity with different tissues. The concentration of the substance, the volume, the temperature, and the impact force. Alkali versus acid. This is a typical question. Which one is worse, or why one is worse than the other. We know traditionally that alkali burns are worse, and the reason is that it’s a lipophilic substance that can easily penetrate into the eye. So it not only damages the ocular surface, but it will damage intraocular structures like the trabecular meshwork, the ciliary body, the lens, so the injury to the eye is much more profound. However, acidic substances can also cause devastating injuries to the ocular surface, but they will cause protein coagulation in the epithelium, and that will act as a barrier for further penetration of the acid. And that is why they’re typically more limited to the ocular surface. What are the clinical findings in the acute setting? Well, the eyelid edema, deepithelialization of the skin and the cornea, loss of eyelashes, eyebrows, there will be corneal and conjunctival epithelial defects, chemosis is common, conjunctival inflammation, a very important factor: Limbal ischemia. The cornea may become edematous and cloudy, and occasionally there can be perforation. So this is a photograph — sorry. Let me go back. All right. So… This is a photograph of a patient with an acute ocular injury. You can see some limbal ischemia. You can see the chemosis, the loss of eyelashes. Actually, these have been cut, probably to try to avoid further trauma. Other clinical findings can include cataract. Sorry. Other clinical findings can include cataract, like in this patient, with an alkali burn. You can see it has a complete epithelial defect, and the alkali has penetrated into the eye, giving the patient glaucoma and a dense cataract. Intraocular pressures are an important complication from injury. It can result in damage or inflammation to the trabecular meshwork. So checking the intraocular pressure in the acute setting is very important. I can’t stress this enough. Oftentimes, patients will be referred to me during the chronic phase, for visual rehabilitation, and all the work that we do in reconstruction and sometimes artificial corneas or keratoplasties have limited benefit, because the optic nerve has been damaged, and nobody cared to properly control the pressure in the acute phase. There can also be secondary damage to the retina and the optic nerve, despite the increase in intraocular pressure. So even if the pressure is not high, the diffusion of the chemical and the intraocular inflammation that is produced can damage the optic nerve. It can even cause retinal detachment. But this has been proven in a mice model, where there’s increased TNF alpha production in the vitreous of patients, of animals, of the chemical injury model, and they see how this produces damage to the optic nerve. So this is the first question of our webinar. What is the most important prognostic factor immediately after ocular chemical injury? You have three options. We’ll give you 30 seconds to answer. All right, very good. So limbal ischemia is by far the most important prognostic factor. So this is one of the things that we really want to look into, when we’re examining these patients in the acute setting. And it helps us, one, determine what kind of treatment this patient will need, and two, helps us have an idea of what the prognosis will be. The reason is: Limbal ischemia can give secondary limbal stem cell deficiency, which will have chronic negative implications in these patients’ prognosis. The classification. Again, because limbal ischemia is so important, of all the classifications that are available for chemical injuries, the common element is the amount of limbal involvement. Because studies have shown that the relative proportion of surviving limbal tissue is actually a major prognostic factor. So this is the Roper-Hall classification. This is perhaps the simplest one. It goes one through four, with good, guarded, and poor prognosis, and it has to do significantly with the amount of limbal ischemia. No amount of limbal ischemia, only corneal epithelial damage, that’s one, between 1/3 and 1/2, that would be guarded prognosis, that’s grade 3, and more than half of the limbus would be poor prognosis with a grade 4. Dua’s classification. Same idea. Just a little bit more complicated. It goes from 1 to 6, but again, it’s still talking about the amount of limbal involvement, and the percentage of conjunctival involvement, and it gives you an analog scale. So our next question is: Based on the Roper-Hall classification, how would you grade this chemical injury. Excellent. So yeah, it is a Roper-Hall 2. Because there is some amount of limbal ischemia, but it’s less than 1/3. So according to Dua’s classification, we’re sticking to the same patient. What is the prognosis, then, for this injury? Excellent. The prognosis is good. Because it is not very good, because there is some limbal ischemia, but the amount of limbal ischemia is small. So the course of the disease. We separate it into different phases. The immediate phase, the emergency, then the acute phase, which is from 0 to 7 days, the early repair phase, when we want to start seeing things healing, between 7 and 21 days, and then the late repair is after 21 days. But when we’re talking about the management, it would be immediate, acute, less than 6 weeks, and then chronic would be over 6 weeks. Management of the immediate phase. So we all know this. We just need to irrigate. At the site of the accident, tap water is okay. It may cause some corneal edema, because it’s hypotonic compared to the corneal stroma. But that’s okay. We’re gonna irrigate. In the hospital setting, we’ll start immediate irrigation, we’ll use anesthetic drops beforehand, we can use a Morgan lens to facilitate irrigation, use at least 2 liters of irrigation solution, check the pH, and continue until we normalize the pH. Always remember there’s correlation between time to irrigation and the final outcomes of these patients. This is just showing you how to use the Morgan lens. I’m sure you’re all familiar with this. We’ll anesthetize the eye first, insert the lens, attach it to our irrigation solution, and then we will continue instilling the fluid through the ocular surface. So the clinical examination of the acute patient. It’s important to check and sweep the fornices to remove any particulate matter. For example, they may have some particles in the fornices, under the eyelids, that continue to release the alkali. So we need to make sure that we examine the ocular surface very thoroughly. And of course, we’ll remove the contact lens. Management of the immediate phase. I always like to see what’s the evidence out there. And the truth is that if we look at all the evidence and all the studies, the choice of the most effective solution is kind of equivocal. There are reports that favor one or another, but in the end, they don’t seem to have a significant impact in the final outcome. So as long as we irrigate, we can irrigate with whatever we have. So the next question is: During the acute phase of a chemical injury, which of the following medications will be most likely to limit further ocular surface damage? Excellent. Controlling inflammation to try to limit further damage to the ocular surface is very important. Sorry about this. Okay. So during the acute phase, the main objectives will be to decrease inflammation, avoid further epithelial and stromal breakdown, and foster reepithelialization. So whatever we need to do to achieve these three goals we will do. And this is just an algorithm to give you an idea. We’re not gonna go through this in detail, but after the immediate irrigation, then you will use corticosteroids to control the inflammation, you’ll try to promote reepithelialization, depending on the severity of the injury, you will need a tenonplasty, you may need serum or platelet rich plasma, possibly transplantation, you’ll take care of the intraocular pressure, you’ll prevent infection with topical antibiotics, and prevention of stromal breakdown, perhaps with frequent lubrication, perhaps with medications like oral doxycycline or vitamin C. So this is our next question. Which corticosteroid timing and regimen is most appropriate after chemical injury? Excellent. So the question is a little tricky. But the idea is that you want to start them quickly after the injury happens. Of course, after we’ve irrigated, and you want to keep them pretty frequently for the first couple of weeks. So the use of corticosteroids will be to control the inflammation. We want to inhibit all these neutrophil infiltrations that will happen after the injury. And again, we do want an intense regimen, particularly in those patients with grade 2 or higher. After the first week, corticosteroids can maybe inhibit epithelialization, can maybe inhibit collagen synthesis, potentially increase the risk of corneal perforation, so we will start tapering them down only if the epithelium has not healed. If we have managed to heal the epithelium within the first week or so, then we don’t really need to lower our dose of steroids, but if we haven’t, we will. We will not taper them off. Just down. And we want to have a lower dose by two weeks. So the antiinflammatory therapy — it is necessary, again, to augment the suppression of the inflammation. They have fewer local side effects. We use systemic corticosteroids. And typically on severe injuries, we can give prednisone, 1 milligram per kilogram, and taper over a few weeks. By giving oral steroids, we can decrease the side effects of the topical steroids. There are some reports for the use of other immunosuppressants that might be needed. I usually use these immunosuppressants more in the chronic phase to chronically suppress the inflammation with a steroid-sparing agent. But they can certainly be used in the acute phase if needed. The prevention of the stromal breakdown is important. It is a major concern, because in the absence of cells to reepithelialize and inflammatory cells like neutrophils to cause necrosis, perforation is a real concern. We can use collagenase inhibitors like tetracyclines, vitamin C, I think those are the two most commonly used. I typically use tetracycline at a relatively higher dose than you would use for your, say, blepharitis or other ocular conditions. The benefits of tetracyclines: They inhibit neutrophil migration. They will suppress the synthesis of oxygen radicals. They will inhibit metalloproteinases in the cornea, and again, you can use doxycycline up to 250 milligrams BID, but the starting dose can be about 100 milligrams BID. You cannot use it in kids younger than 8 years old, and of course not in pregnant women. The promotion of the reepithelialization, it’s important. This is one of the main goals of the acute phase. Once the epithelium heals, we feel that a lot of the risks go down. So very frequent preservative-free lubricants, we’ll use antibiotic drops until the surface heals, and you want to avoid those that have significant techniques to the ocular surface. Bandage contact lenses are useful, like silicone hydrogel contact lenses, and sometimes on epithelial defects that are persistent, we can try large diameter gas permeable lenses, or the PROSE scleral lens, with very good results. Amniotic membrane transplantation in those patients that are not healing — then we need to step it up and go forward. So we can use amniotic membrane transplantation in the form of a ProKera ring, a symblepharon ring that has the amniotic membrane attached. It’s very convenient and can be placed in the office, or we can take the patient to the operating room and actually transplant the amniotic membrane onto the ocular surface, glue it and suture it to promote reepithelialization, and typically we’ll need to do this in patients who have injuries grade 3 or above, but sometimes in injuries that are grade 2, if they’re not healing, we’ll have to do it. Serum tears. We use frequently, and also plasma tears. And they both have autologous growth factors that can help with epithelialization. There are many studies showing their effect in the help of treatment of dry eye, and healing of persistent epithelial defects. But we really don’t have any good long-term or prospective studies looking at the specific effect on chemical burn. They certainly don’t have a downside, and I typically will use them. The tarsorrhaphy — we wouldn’t do a tarsorrhaphy right away. But after it’s been a couple weeks, if the surface is still not healing, these may be necessary to give further protection, and continue to promote epithelialization. So our next question is: Which one of the following pitfalls is more serious during the management of chemical injury patients during the acute phase? Perfect. And it will be — however we decide to heal the surface, it’s okay, and we have many tools to do that. But if we do not follow the intraocular pressure, all our effort to heal these patients will not be translated into a successful visual rehabilitation. So as we talked about, there’s many factors causing glaucoma, but particularly the chemical that diffuses into the eye and damages the trabecular meshwork is very easily overlooked. Because we don’t want to be putting anesthetics on the surface that’s already compromised and touching the surface and checking the pressure every time the patient comes — but it is a very important thing that we need to do. Tenonplasty is another technique that can be used during the acute phase, and we reserve it mostly for severe injuries with loss of limbal vascularity. And the idea is that we will advance Tenon’s to try to bring vascular supply to the limbus, and get ahead of the game, trying to avoid anterior segment necrosis, scleral ischemia, melting, and scleral ulceration. Down here, this is a photograph I took from a paper that was recently published this month in Ocular Surface. But it was helpful to illustrate how this person who has some limbal ischemia — we could try to see and promote healing with conservative measures, but these patients with more extensive limbal ischemia, their group is a high proponent of tenonplasty to improve vascular supply in this set of patients. Again from their paper, I wanted to show you the results of the patients with limbal ischemia, the advanced Tenon’s, and when we do the angiography, you can see the vascularization that was brought to the limbus. So again, evidence-based recommendations for the management of the acute phase. Again, it’s all based on retrospective or small prospective cohort studies. However, we do know that corticosteroids are helpful. That tetracyclines might be helpful, but the literature is limited. Bandage contact lenses and amniotic membrane. Mostly the use of amniotic membrane — we haven’t shown that it changes the outcomes, but it can help in speeding up the epithelialization of the ocular surface. Again, tenonplasty, we only have observational studies. And the treatment of the high intraocular pressure, we’re extrapolating obviously from cohort studies, with the outcomes varying by IOP. Next question: Which one is the last surgical intervention that should be performed in a patient during the chronic phase of ocular chemical injury? So the last intervention that we need to do is keratoplasty. Otherwise keratoplasty will have very little chance of succeeding, if we have not managed glaucoma, corrected eyelid abnormalities, or managed the ocular surface. So we’re getting into the chronic phase. We’ve gotten the patient through the acute phase, we’ve managed to heal the ocular surface, and now we’re left with the sequelae of the chemical injury, and we need to visually rehabilitate these patients. Know from the start that these will require a multidisciplinary approach. That it will be the cornea surgeons, oculoplastic surgeons, glaucoma specialists, all working together. Because the effects of the chemical injury can be very diverse, and will need a team approach for a successful outcome. So our first goal would be to try to restore a normal ocular surface and anatomy. And obviously visual function. The typical order of the procedures that we will perform will be, first: We’ll correct any eyelid abnormalities, because eyelid abnormalities in the form of trichiasis, entropion, symblepharon will preclude the success of any limbal transplantation, for example, and any other forms of keratoplasty. We’ll make sure we’re managing glaucoma. So perhaps the patient needs a shunt, and perhaps we can do that before we need anything else. Because we know that if they already have glaucoma, and we do a keratoplasty or a keratolimbal autograft, the glaucoma will only get worse with the restriction of the angle, so this is very important. The reconstruction of the ocular surface is next. This can involve amniotic membrane transplantation, buccal-mucosal graft transplantation, and then the keratoplasty, either with natural or artificial cornea. This is the management algorithm, and this is important to understand. First of all, we need to determine: Is this chemical injury unilateral or bilateral? And that’s extremely important, because that will determine what options we have for the treatment. The next important thing that we need to determine is: Is this patient otherwise healthy? Is he a young patient or an older patient? Can this patient have systemic immunosuppression or not? That’s another important question to answer. So for the fornix reconstruction, initially, after the acute severe injury, we can have loss of eyelid tissue, we can have contractures, we can have symblepharon. All of these will lead to corneal complications if it’s not treated, in the form of corneal opacification, ulceration, and even perforation. So all eyelid and fornix abnormalities we want to correct before any limbal or corneal surgery, like I said before. But we want to still delay surgery on the eyelids until the eye is quiet. Operating on a hot eye is never a good idea, and chemical injuries are not an exception. There’s different techniques for ocular fornix reconstruction, and they have to do with the severity of the symblepharon, for example. And we basically define symblepharon, ankyloblepharon, are a result of conjunctival deficiency. So we need to know how much good and viable conjunctiva does this patient have to select the appropriate technique for the fornix reconstruction. So when there’s mild to moderate symblepharon, we can get away with releasing the scar tissue, reflecting the viable conjunctiva, to the palpebral side, and then resurfacing with amniotic membrane. We can add the use of mitomycin C, of 5-FU, to decrease further fibrous response, fibrotic response. However, when the patient has more severe — moderate to severe symblepharon, like this patient here, that already failed the reconstruction with amniotic membrane, what do we do next? Well, we turn to other sources of epithelial cells. And the best next will be a buccal mucosal graft. Again, when there’s extensive symblepharon and ankyloblepharon — the only caveat here is that we have unilateral injury, and the patient is amenable to taking conjunctival graft from the other eye, it can be used, but most of these injuries, even if asymmetric, tend to be bilateral. So we’ll graft a buccal mucous graft, put it on the tarsal conjunctiva, and these will provide epithelial cells to resurface the ocular surface. Management of glaucoma, again, in the chronic phase — this is still important. And we can do shunts, even in patients that have keratolimbal allografts, in patients that have keratoprostheses, for example, in the right side of the slide, but we can only do shunts if we have good enough conjunctiva. And sometimes if the patient didn’t need ocular surface reconstruction, perhaps they do have good conjunctiva for a shunt, but when there’s a lot of scarring of the conjunctiva, GDIs are high risk, and sometimes not even possible, and we have to turn to perhaps ciliary body ablation for the intraocular pressure control. Of course, medical treatment is always an option, and we’re talking for when medical treatment is failing. However, we also have to consider that topical drops have deleterious effects on the ocular surface, and we’re already starting with a compromised ocular surface, so oftentimes we’ll consider surgical interventions earlier than in other patients without chemical injury. The result of limbal ischemia and all this surface inflammation is that the damage to the limbal niche and limbal stem cell deficiency. It will result in corneal-conjunctivalization, neovascularization, persistent epithelial defects and ulceration, corneal scarring, and severe visual loss. Once we have limbal stem cell deficiency, we cannot do keratoplasty alone, because we know this will fail. So it’s actually contraindicated to do. How do we diagnose limbal stem cell deficiency? These are cases of milder forms of limbal stem cell deficiency. We’ll see dull opaque epithelium, in a whorl pattern like you’re seeing in these photographs. When you put fluorescein in the eye, there’s not an immediate stain, but more of a late uptake. This fluorescein is diffusing through the conjunctival epithelial cells that are different from the corneal cells. They don’t have those tight junctions to prevent fluorescein diffusion. We can do impression cytology, we can do confocal microscopy, for further diagnosis. The management of limbal stem cell deficiency, like with any other disease, depends on the severity, and depends on whether it’s unilateral or bilateral. The patient in the photograph here has a failed keratolimbal allograft. You can see the raised limbus, where the allograft tissue was placed. So if the limbal stem cell deficiency is partial, we can use non-preserved lubrication, autologous serum eye drops, topical vitamin A and steroids, and perhaps reverse and get the healthy limbus to take over. When there’s central cornea involvement, we may have to scrape the epithelium and promote healing of the healthier areas of the limbus to try to cover the area of conjunctival epithelial cells, and we may have to do this repeatedly over time, until we achieve the result that we want. We can combine this with amniotic membrane transplantation, and if all of the above fail, we can always do an ipsilateral limbal transplant in the form of a simple limbal epithelial transplantation of conjunctival limbal autologous transplantation. But when the limbal stem cell deficiency is complete, then we have to resort to other techniques. And again, it depends whether we have unilateral disease or bilateral disease. For unilateral disease, it’s easier. We can go to the other eye, and harvest a conjunctival-limbal autograft. We have the option to do cultivated limbal epithelial transplantation, although everything that’s cultivated requires a laboratory for cell expansion. This is very expensive, and it’s not readily available everywhere. So these procedures are only done in certain centers around the world. They’re really not available to most of us. And the other option would be a SLET, a simple limbal epithelial transplantation from the other eye. When it’s complete bilateral injury, these are the patients that are most challenging. We can do a living related conjunctival limbal autograft. This will require systemic autosuppression, so the patient has to be young and healthy. We can do a KLAL from a cadaveric donor. The same goes. Systemic immunosuppression. Has to be a healthy patient. We can do Allo-SLET, but it will likely reject in this setting, and isn’t likely that it will provide clarity and be successful in a patient with bilateral total disease. COMET is another option, but again, the same as CLET, it requires a laboratory. It is expensive. And the visual results obtained with COMET, because the source of the epithelium now — it’s not cornea. It is oral mucosa. And then the visual results are typically not that great. And finally, for patients with bilateral total disease, we have the option of performing a keratoprosthesis. So if we’re deciding on limbal stem cell transplantation, it is not recommended during active inflammation. This is very, very important. And everything has to be fixed first. We need to fix the eyelids. We need to fix the ocular surface. We need to make sure the glaucoma is controlled. So the CLAU technique — we’re gonna harvest, again, a conjunctival-limbal transplant, and the idea is we’re providing not just corneal limbal stem cells, but also some conjunctiva, in case that there’s also conjunctival deficiency in these patients. And then we will transplant it to the fellow eye. In simple limbal epithelial transplantation, we’re gonna take — this technique was born basically because there’s always concern… Okay, if I have one eye that’s damaged, and my other eye is relatively healthy, and I’m gonna take a big proportion of it, to transplant it to my damaged eye, I’m putting my good eye at risk. So if we only take little pieces, and we transplant this, and these will expand, the results are actually very promising, that the ocular surface can be restored this way. However, it is not as effective as a complete limbal transplantation, or a larger portion of limbal transplantation, like we would do when we do, for example, a keratolimbal allograft. But the benefit of being an autologous tissue is very important. So perhaps once we do a SLET, the whole ocular surface won’t be perfect, but as long as we obtain corneal epithelial cells in the center, hopefully the vision will improve. The living related conjunctival allograft. So we need a living related donor that’s compatible. And the downside is it requires performing surgery on two patients, the donor and the recipient. But basically the technique is the same as the conjunctival limbal autograft. And then the keratolimbal allograft. This involves transplantation of limbal stem cells from a cadaveric donor, and we take three lenticules to encircle the entire limbus, 360 degrees, and these patients will require immunosuppression, and the immunosuppression that we use is very similar to that used in patients with kidney transplantation. So this is a quick video that I wanted to show you. This is how you would harvest a keratolimbal allograft from a cadaveric donor. And this technique I learned from Dr. Ali Djalilian, where we use cyanoacrylate, which avoids assistants having to hold the tissue for you. We’ll go right up to the limbus, then use the crescent blade to harvest the limbus in the area of the cornea, and then finally we’ll remove the tissue that we harvested, and place it in the patient’s eye, and we’ll do three like this, to make sure that we can encircle 360 degrees around the patient’s limbus. So once we finish harvesting the tissue, we’re gonna move on to our next video here. Then we’re gonna put fibrin glue. We’re gonna recess the conjunctiva back. The patient’s conjunctiva. And we’ll put fibrin glue, and then we’ll glue all these pieces onto the limbus. Typically we’ll scrape the corneal epithelium. We can use amniotic membrane over a contact lens. And we’re gonna skip that video. This is limbal stem cell transplantation again. Requires immunosuppression. Similar to that used by patients with kidney transplants. So it will be oral steroids that will be short-term. Tacrolimus or cyclosporine. And mycophenolate or azathioprine. When we think about these procedures, we have to think about a patient that is young. Typically younger than 55 years old, and that is relatively healthy, able to tolerate systemic immunosuppression. What are the complications we can have with these procedures? Immunologic rejection, chronic ocular surface exposure, and these complications can lead to ocular surface epithelial breakdown, thinning, and progressive corneal conjunctivalization. And glaucoma is also a complication of limbal transplantation. It has to do perhaps with the changing of the angle, with transplantation of the tissue, but perhaps also with the frequent use of steroids. So our next question: To the audience. Which of the following is not an important factor in preventing complications after limbal stem cell transplant? Right. So B is clearly wrong, because we know the keratoplasty will fail and have complications unless we properly reconstructed the surface. With a limbal transplant. So other complications: Most important factors in preventing complications after a limbal transplant include making sure that the tear film is good. A full correction of all adnexal abnormalities, proper handling and dissection of limbal grafts, and adequate immunosuppression. Studies are showing that immunosuppression after keratolimbal allograft has to go beyond two years to prevent late rejection. In this photo you’re seeing a patient with acute rejection of a keratolimbal allograft. Corneal transplantation. So perhaps we’re lucky, and just with the limbal transplantation, we’re able to resurface the corneal epithelium. And if the injury wasn’t that severe, and it did not cause corneal scarring, we can get away without a keratoplasty. But a lot of patients will require an optical procedure after the ocular surface has been restored. And this can be in the form of a penetrating keratoplasty, a deep anterior lamellar keratoplasty, and if that is not successful, it may require a keratoprosthesis. Keratoprosthesis surgery — we do it when there’s limbal stem cell failure. It can be unilateral or bilateral chemical injury, and the most commonly used right now is the Boston type 1 keratoprosthesis. So our next question: Getting into artificial corneas. Which of the following is an ideal case for a Boston type 1 keratoprosthesis placement? So if we’re starting to build the algorithm in our minds, the best candidate will be an older patient, because older patients are not susceptible to immunosuppression, and bilateral, because if we have unilateral disease, then we can source the other eye for cells to restore the limbus. But when the injury is bilateral, and it’s an older patient with comorbidities, we know we cannot do immunosuppression, so we know the only option is actually a keratoprosthesis. So when we’re evaluating somebody for a K-Pro, we need to make sure they’re amenable to long-term risks, they’re gonna be able to follow up regularly with us, they’re gonna need to be on antibiotic prophylaxis daily, and other chronic maintenance issues that happen during their follow-up visits, that we need to attend to. So we need to make sure that patients are reliable and will be able to follow up. Because there can be complications, and those can be frequent sometimes. Retroprosthetic membranes, glaucoma is a big issue in patients with keratoprosthesis, as it is also in patients with chemical injury. Sterile corneal necrosis can happen, infections, retinal detachments, et cetera. Glaucoma is of particular interest, because we already know that we’re dealing with a susceptible nerve that has sustained damage from the chemical injury, perhaps, and now we’re gonna put a keratoprosthesis, which most studies are showing — that the risk of developing glaucoma or progressing glaucoma after a keratoprosthesis is high. So we need to have a plan, how we’re gonna manage glaucoma in patients with keratoprosthesis. And in my mind, a tube shunt before K-Pro or at the time of K-Pro is highly recommended. But now we’re dealing with severe chemical injuries, and some of these patients may not actually be amenable to placing a shunt. They may not have good enough conjunctiva. I’m gonna share with you this video that we’ve done with our team. This is a patient, a young male, with very severe burn to both eyes. And now this is just a review. We have mild symblepharon. We can just reflect the good conjunctiva and cover with amniotic membrane. When it’s moderate, we may need bolster sutures, but when it’s severe, we’ll need another source of cells. This is a patient with mild symblepharon. We’re recessing the conjunctiva, anchoring to the superior eyelid with bolster sutures, and we’re gonna resurface the rest with amniotic membrane. And she actually did very well after this reconstruction, without recurrence of the symblepharon. And you can see here we’re just adjusting the amniotic membrane. We typically use fibrin glue for all of these procedures without the use of any sutures except the bolster sutures. Now we just want to trim it. So it’s important for you to assess before surgery how much viable conjunctiva these patients will have. So at the end, once we’ve reflected the conjunctiva, we’re gonna put our bolster sutures, and we leave those in place for at least a couple of weeks until everything is healed and anchored in place. But in cases of severe symblepharon or moderate to severe symblepharon, like in these patients, then we may need to resort to buccal mucosal grafts. So we’ll harvest it from the lip. We’ll take as large of a graft as possible, and we’re gonna put it onto the superior and inferior eyelids, anchor it with bolster sutures, and then these patients will typically require some form of immunosuppression systemically during this procedure, to decrease the fibrovascular response, and then after we’ve reconstructed, if we’re successful, then sometimes in these patients, we’re actually able to perform a tube shunt under the reconstructed conjunctiva, in combination with an artificial cornea. So in these patients, we didn’t do a limbal transplantation, because the damage was so severe that we knew it wouldn’t be successful. We like doing these combined procedures. We like to put the shunt in the pars plana, so we’re gonna do a pars plana vitrectomy, we’re gonna shave the vitreous base, and we’re gonna place the shunt 4 millimeters posterior to the limbus, and we’re suturing the artificial cornea in place, and you can see how this newly reconstructed conjunctiva can give us a nice closure for that Baerveldt shunt, and this way we have a good plan for glaucoma in this patient that, once we looked at the optic nerve, we could already see the nerve was pale and damaged from glaucoma. But indeed he did regain some vision. Unfortunately this patient had a retinal detachment before this surgery, secondary to his injury, on top of the optic nerve damage. That will limit his vision to about 20/400. And that’s why you see those silk sutures, traction sutures, on the muscles. That’s because we had to perform retinal surgery to repair the retinal detachment at the same time of this surgery. So the final outcomes for this small series that was just published in Ocular Surface. You can see the patient prereconstruction and after the surgery. This is the same patient, again, with moderate symblepharon, and the third patient with mild symblepharon, that had a beautiful outcome. And this is our last patient with moderate symblepharon. This patient actually was the only one in the series that the shunt didn’t work postoperatively, and we needed to perform laser to control the pressure. When the Boston type 1 is not an option, then we have to resort to type 2 keratoprosthesis. These are a more infrequent type of keratoprosthesis, and they are performed in patients where there is severe dryness, keratinization, no tear production, and there might be loss of eyelid, no good blinking function, and we have two options. The type 2 Boston keratoprosthesis, which is similar to the type 1, except that it has a longer stem that will go through the closed eyelids, and the second option, which is the favored one, is the osteo-odonto keratoprosthesis, which is done with the patient’s tooth as the anchoring haptic for the PMMA cylinder, and a buccal-mucosal graft will go over this tooth, and this surgery is performed in two steps. It’s a very lengthy and technically challenging surgery. And it’s only really done in a few centers around the world. So again, evidence-based recommendations. We don’t have a good randomized controlled clinical study that will tell us this is the very best thing you can do for your grade 1, 2, 3, and 4 chemical injury, in the chronic phase. But we do know that keratoprostheses are successful, that keratolimbal allografts work, that SLET has good outcomes as well, and when surgical intervention is needed, and for the fornix reconstruction, we know that amniotic membrane and buccal-mucosal graft can be successful, and antimetabolites may be needed, as well as systemic immunosuppression. And there are some observational studies that suggest the benefit of tube shunt if adequate conjunctiva is present, or cyclodestructive procedures for the treatment of glaucoma. So new ideas for severe chemical injuries. We’re doing okay, but we’re not doing great. We definitely need better things to manage these patients. The main goal is: We need to bring some new viable cells that can resurface the ocular surface. So allo-SLET is an idea. I learned of this on my trip to India. Last year, to the service in Chennai. And their group is doing — during the acute phase, they perform Allo-SLET from a cadaveric donor. And these help bring a source of epithelium to speed up the ocular surface epithelialization. And interestingly, this is a photograph from their study, showing how the initial chemical injury shows and then after the allo-SLET, how the epithelial defect resolves. It is possible — the way that this is done is you’ll harvest your little SLET from a cadaveric donor, and put them sort of all over the ocular surface. You can glue them and put amniotic membrane over that. So I wanted to acknowledge Medi Eslani and Ali Djalilian and Dr. Faris Karas for helping me with slide preparation. And this is all I have for you today. Thank you.

So the question is — let me start from the beginning. How to accurately measure IOP in such a damaged surface. Of course, always having an accurate measurement of the IOP is difficult. So on patients that I would recommend perhaps using the Tono-Pen or the pneumotonometry, if available, but sometimes I understand these can be challenging. On patients that have keratoprosthesis, this is also a problem. And we can’t applanate these patients. We can use a pneumotonometer in the sclera, and it gives us a range, kind of an estimate, but most of the time finger palpation is what we have to resort to. And this is a big problem in managing glaucoma in patients with keratoprosthesis, because we don’t necessarily have an accurate measurement of the pressure that we can more precisely titrate the treatment to. So how long do we have to irrigate at the beginning before we start the treatment? And should we irrigate the next days too? No, I think the irrigation is only on the first initial encounter, and we have to irrigate until we can normalize the pH. What is your experience in scleral lens use in the acute and late phase of severe ocular burns? We’re very lucky here in our service that we have a large contact lens service that does scleral lenses very well, including the PROSE lens. And I think for the acute phase, we use it on non-healing epithelial defects. We have been very successful in general with the use of the PROSE lens, with daily insertion and removal for the treatment of non-healing ulcers. In the chronic phase, we have found that scleral lenses — even if the cornea is irregular, and may be somewhat vascularized, they can improve vision as well, and protect the ocular surface from further damage, from abnormal eyelids. So we actually use them quite frequently. Any role of stem cell transplantation in acute presentation in which there is total stem cell loss, both corneal and conjunctival? So yeah. I think the only role would be this new idea of the allo-SLET, where we use — I mean, we don’t want to waste, if there are any autocells for later reconstruction, but we can use a cadaveric donor, and put these little pieces of limbus and conjunctiva over the ocular surface, cover it with amniotic membrane, to provide a source of epithelium to speed up the healing process. Even if the injury has low grade, is it advised to maintain the corticosteroids for a week or two? The answer is yes, because you really want to minimize the inflammation. The steroids are gonna be the most important in your patients that have higher grade injuries, but even if the injury is low grade, if there’s evidence of ocular surface inflammation, I think there is a role for topical steroids. Which type of suture is more suitable and less inflammatory reaction developing? In general, when we resurface the ocular surface, we try not to use sutures. The only sutures that we use are the ones that anchor to the eyelid with the bolsters, to anchor the tissue to the tarsal aspect of the fornix. But if you need to use sutures because fibrin glue is not available, I think nylon probably is the best one. How long does keratoprosthesis last? Does it need replacement? So ideally we would hope that it would last for a long time. We have patients with keratoprosthesis and chemical burn that have retained the keratoprosthesis as long as 10 years, or they still have them today. In our service, sometimes the keratoprosthesis needs to be replaced. If there are complications like, for example, exposure and melt. That would be one reason to have to replace the keratoprosthesis. Is there a place for anti-VEGF management of corneal neovascularization? So I think that anti-VEGF can be effective in the acute neovascularization. But as long as we haven’t reverted the ischemic drive for the vascularization, I don’t think that using anti-VEGF will be successful. So both have to be combined with a strategy to reduce the ischemia. What is the type of glue that you’re using for AMT? It’s fibrin glue. Do we need to remove pseudomembranes in the acute phase? I think debridement is a good idea, just like we do in patients with Stevens-Johnsons. With a chemical burn, removing the debris and cleaning the ocular surface, I think, can be important in the acute phase. What do you advise to manage pain? This is a very tricky question. Not a tricky question, but this is a difficult subject. I think sometimes bandage contact lenses, tarsorrhaphies, amniotic membrane, and all of these sort of bandage treatments to the ocular surface can help with pain, and then we typically use oral medications. Since amniotic membrane is not available in my country, do you advise using dry amniotic membrane? Yes, if that’s the only thing you have available, I don’t think it’s a bad idea. The dry amniotic membrane once you hydrate it — can be used just the same as you use the other amniotic membrane. The dry amniotic membrane that we get here is typically a rounded disc. So it wouldn’t be helpful to reconstruct the fornix, but definitely it’s useful to place over the cornea, to promote healing. It’s actually very easily done at the slit lamp, in the clinic, and you’ll put a bandage contact lens over it, to keep it in place. What is the proper timing of limbal stem cell transplantation? So the proper timing for transplantation is when the ocular surface inflammation is resolved or controlled with medications. That is the only time that we can think about limbal transplantation with the hope of success. What is your management of corneal melting following chemical injury? What about descemetocele or perforation? Unfortunately in the acute phase, these are feared complications that could happen. Corneal melting can be treated in many ways. We can use cyanoacrylate glue, and some small studies have shown that that can stop the melt, or reduce the melt, and prevent the perforation. Once the perforation is there, if the perforation is small, we can use an amniotic membrane pack for the descemetocele, but if it is larger, we might have to do lamellar keratoplasty, which can have a high failure rate in the acute phase, but at this point the purpose of doing the transplantation is a tectonic one, to preserve the globe, and later we’ll worry about visual rehabilitation. You can add doxycycline, medroxyprogesterone, and things that will decrease the risk as well orally. So do the buccal and labial mucosa have stem cells? So the answer is yes. They’re just not — they won’t transform into corneal epithelium. They’re mucosal epithelium. And so they’re not gonna form these transparent epithelium that can be optically favorable for improvement of vision. And that’s why the COMET procedure, where we cultivate oral mucosa, and then we transplant a sheet of this into the ocular surface, can promote healing of the ocular surface, and improve some of the abnormalities, but visual outcomes are typically not great. Okay. I think that’s the end of the questions.

Download Slides


February 22, 2019

Last Updated: October 31, 2022

Leave a Comment