Lecture: Postoperative Management of Penetrating Keratoplasty

In penetrating keratoplasty, the surgery is only the beginning! Postoperative management is critical in ensuring good results and happy patients. This webinar will focus on immediate postoperative issues like epithelialization and glaucoma as well as longer term issues like rejection and astigmatism management. Log on to hear frequent visiting faculty Dr. James Lehmann share his tips and wisdom.

Lecturer: Dr. James Lehmann, former Orbis staff ophthalmologist (2005) and currently a cornea surgeon at Focal Point Vision in San Antonio, Texas, USA


DR LEHMANN: Hey, everybody. Good morning. This is Dr. James Lehmann, coming to you from San Antonio, Texas. It’s a little chilly here. I’m gonna share my screen, and we’ll start the talk about post-op management of penetrating keratoplasty. Good morning, everybody. It’s about 8 am here in Texas. We’re gonna be talking about PKs today, and we’ll get started. So a little bit about myself. I’m an ophthalmologist in private practice in San Antonio, Texas. That’s down here. I do cataract, cornea, and refractive. I don’t have any relevant financial disclosures for this talk, and Dr. Tony Aldave has given some of the photos for this presentation, so I’m giving him thanks before I start. I’ve had the opportunity to work with Orbis and SightLife over the last decade, traveling to different countries, and I’ve done this same talk probably three times in the last year. Overseas in Myanmar and in India. So we’ll get going! So the objectives today are gonna be to be able to recognize and manage post-PK complications. We’re also going to learn how to prevent those complications through proper intraoperative steps, and we’re gonna learn to recognize and treat rejection. I think we can achieve those goals. We’re gonna start with some questions. So the first question is: Which of the following is not a medicine used to treat acute graft rejection? Topical dexamethasone, sub-Tenon’s triamcinolone, topical cyclosporine, or oral prednisone? So go ahead and submit your response, and Lawrence will tally those up and show us the result here. Okay. We can go ahead and see the results. Most of you answered that topical cyclosporine was not used to treat acute rejection. That’s the correct result. And if you have a defect, which is not a modality to treat that defect? Soft contact lens, Latanoprost drops, amniotic membrane, or tarsorrhaphy? Which is not helping to treat the defect? Go ahead and answer the questions. I’ll sip my tea. And you can show the results there, Lawrence. And last but not least, after successful pK surgery in an adult, one should begin removing sutures. Go ahead and submit your answers and you can go ahead and show shows, Lawrence. Okay. All right. So we’ll look at those again at the end and see how we do. So this is what PK looks like. The end result is a clear visual axis with a good shape. That’s what we’re aiming for. In the United States, if you look at this graph, over the last 15 years, the number of PKs has plummeted. Now we do more endothelial keratoplasty than penetrating keratoplasty, and the reason for that is because in the United States the most common reasons for grafts are Fuchs’s dystrophy, so we’re doing DMEK and DSAEK for those. The most common reason for keratoplasty in the US is keratoconus or repeat corneal transplant. That’s very different from other countries. In India, the number one reason for penetrating keratoplasty is gonna be a therapeutic graft for ulcers. I’m gonna move a little bit. The sunlight is hitting me right there. There we go. And I looked it up, and in Colombia, it would be bullous keratopathy and then therapeutic. In the US it’s mostly keratoconus and repeat grafts. What are complications to corneal transplantation? These are people to avoid from the get-go, so you don’t have those problems, like non-healing epithelial defects, dry eye, suture infections. So if the ocular surface is uncontrolled, either from severe dryness or something very bad like Stephen Johnson’s or OCP, something more common like limbal stem cell deficiency, or bad exposure, those things have to be fixed first before you do a penetrating keratoplasty. Or you have to choose a different modality like a keratoprosthesis. Again, the best way to avoid complications is to not do a penetrating keratoplasty in a patient that’s not a good candidate. Other contraindications include active infection, unless you’re trying to treat that infection, and in certain situations, if the HSV or VZV is active, you need to treat it first with Valtrex and get the disease quiet prior to surgery. Uncontrolled glaucoma is another contraindication. If you do a graft in somebody with bad glaucoma, the graft is gonna fail and they’re gonna need a tube surgery, and that’s gonna really cause the graft to fail. So the glaucoma has to be fixed before you do the penetrating keratoplasty. Also, multiple previous rejections. If you’re gonna do the fifth transplant on somebody, you’ve got to figure out why the first four didn’t work and decide if you can still do another graft or if they need something like a K pro. And lastly, the inability to care for the graft. If somebody lives ten hours away from your clinic and they need a PK, guess what? If they don’t come back and see you, they may end up worse than they were before. We don’t do a graft in somebody who can’t come back for follow-up care. Again, this is kind of the result we’re looking for. And this is the typical pattern I do on my transplant. It’s about an 8 millimeter graft. Okay? And then it has 12 interrupted sutures, like this. And then a 12-by running suture. So we’re gonna get into suture combinations and which one may be the best one. But I have gravitated to this over the years, because it gives me a better flexibility, in terms of taking the sutures out early and wound stability. All right. So we’re gonna just jump right into it, and then we’re gonna talk about the complications. But on the first post-op day, what are you looking for? You’re looking for pressure, you’re looking for wound leak, and you’re looking for patient comfort. You don’t really care too much about the vision at this point. Obviously people aren’t gonna see really well, counting fingers. You don’t want them to be hand motion. That would be bad. But counting fingers is pretty typical. You want to do a Seidel test along the graft, touch on the eye a little bit with a finger or a Q-Tip, and you want to check the pressure. If the graft is very clear on the first day and there’s no epithelium, it suggests that the pressure is kind of high. If there’s no epithelium you won’t get epithelial edema and you’ll end up with a very clear graft. The patient may be seeing well and you’re patting yourself on the back and the pressure could be 40, so you have to check it. We had a question submitted from a doctor from Myanmar. I was just in Myanmar, so shoutout to Dr. Sandar Khiao. I may not have said that right. But which tonometry is best to measure post-op IOP? If you can use Goldmann tonometry, that’s great, but you often can’t do that early in the postoperative course, because the cornea doesn’t have a good shape and you’re not getting good measurements or the cornea is very sensitive. We use the Tonopen in our clinic, or you could use a pneumatonometer as well. A study in New Zealand showed that the Goldmann, the Tonopen, and the Pascal Dynamic tonometer did the best. I looked at this Pascal Dynamic, and it’s good, but it may be hard to do on graft patients as well. The Tonopen might be good as well, and the pneumatonometers, which are pretty cheap. That’s the best way to measure post-op IOP. And what medicines do we use? You use a topical antibiotic from day of surgery until the epithelium heals. You have to look at your area, find out what’s affordable and also has good coverage. Fluoroquinolone is fine. If you’re in a hospital-based environment, you can look at what are the sensitivities to common pathogens and taper to that, but you can’t go wrong with a fluoroquinolone like moxifloxacin or gatifloxacin. Now a steroid. We’ll get into a little more details on the steroid. You start out hitting them pretty hard and decrease it, ending up one time a day for life. Again, here are my steroid recommendations. I use Durezol in the short term because it’s more convenient dosing, but you may not have that available in your country. We use it three times a day, which is the same as six times a day on prednisolone, and you taper monthly according to the inflammation, and check the pressure. If they’re on Durezol and you bring them back in three weeks and the pressure is high, obviously you need to decrease that and put them on an ocular hypotensive. In the long term, I keep them on Pred Forte, every other day, or Lotemax or FML every day. This is for life. A lot of places I visit, they ask: Does a patient need to be on a medicine for life? There’s an article in Ophthalmology in 2009 that looked at that, and it’s been shown that a patient on steroids is gonna have a lower chance of rejection episodes. Once somebody is on a stable graft, I see them every six months, keep them on steroid for life, and every visit I check the pressure. If they have signs of glaucoma, I put them on a drop like timolol and follow visual fields like a glaucoma suspect. But if it were me, I would want to have a steroid for life. We also had this question submitted from India, and it was: In steroid responders, what steroids do you prefer? Say they’re on prednisolone or Durezol right after surgery and they come back in a month and their pressure is 30. A good thing to do at that time would be to put them on fluorometholone or loteprednol, and also Combigan or timolol, something like that. These medicines, as we all know, they’re strong steroids but they’re not absorbed through an intact epithelium. They work great in ocular surface issues, transplants, and stuff. Okay. Now we’re at post-op week one. What are you looking for at the post-op week one visit? The main thing is to see the epithelium has healed. It obviously hasn’t healed in this patient. If it hasn’t healed, you need to go through an algorithm to fix it, which begins with a bandage contact lens, BCL, AMT, and tarsorrhaphy. You also have to consider putting them on oral Valtrex. Not quite at first, but if they don’t heal with one of the first line modalities, putting them on Valtrex also helps if there’s any kind of herpetic process going on. You also want to check the pressure, of course, like we just discussed. There shouldn’t be a wound leak at this point, but there could be if the epithelium hasn’t healed, and by this time the patient is pretty comfortable, at a week. Most of the time they just have discomfort for two to three days. And we see them in a month. What happens at a month? We’re looking at pressure, mainly. You want to see the graft is getting a little clear. If edema hasn’t cleared by a month, you have to start thinking about primary graft failure. If it’s really bad edema, it’s an easy decision to come to, but if it’s only mild edema, you can give it a little more time, perhaps do a cell count. All right. So let’s talk about postoperative complications. We’re gonna start with wound leak, and then there’s Urrets-Zavalia syndrome, persistent mydriasis after surgery, persistent epithelial defect, primary graft failure, infection, suture related problems, and how to identify and treat rejection. Starting with a wound leak, you would see the patient on day one, and you would put a Seidel in the lower right here and see they have leakage. Okay? The eye will be soft at this point, and to fix this, if it’s a suture tract leak, it’s just leaking from a suture tract, you can put a contact lens on the patient. Once it epithelializes, it will work. But if there’s a wound gape, you may have to put a new suture. That gets to suturing. How do you like to suture? We’re looking at 90% depth on the suture like you see here. We don’t want to go 100%, because you’re gonna have more wound leaks and everything. And you have to have good apposition of tissue. Not too loose, and to have enough of them, frankly. So if you’re doing just purely interrupted sutures, you have to do 16. And if it’s a wider graft, you have to do more. Here is the Urrets-Zavalia syndrome, a patient with a pupil dilated. Let’s say you didn’t do any cataract surgery or anything at the time that they present. Most of the time, this is from very high pressure, and you get iris death from the high pressure. We’ve seen this more commonly in DSAEK surgery, but it can happen in PKs as well. In the old days, they used to do PIs on every patient that had penetrating keratoplasty. I don’t do that routinely. And you’re checking the patient the day after surgery. This is a pretty rare complication. There’s no real treatment for it, other than… I’ll go back to that. No real treatment for this, but now there’s an artificial iris available. Obviously you want to treat the pressure. But most of the time, if the pupil is blown, it’s gonna stay blown. Okay. This is a much more common problem and it’s a persistent epithelial defect. You can see that this eye is kind of sick to begin with. Right? They have vessels leading up to the graft. This is probably a couple weeks after the surgery, and they have this kind of necrotic, ugly-looking stroma right there. This would be an unusual case, but let’s just say the epithelium is not healing. You’ve got to figure out: Why is that the case? So you look at their underlying ocular surface disease. Do they have exposure? Do they have dry eye? Do they have something bad, like limbal stem cell deficiency? Or OCP? Well, all those things have to be addressed before the transplant. Or at the same time. With some sort of lateral tarsorrhaphy. But let’s say they had a normal ocular surface and they still got it. You’ve got to make sure they’ve got good nutrition. Diabetes is rare but sometimes can cause problems. Exposure, and then HSV, like we talked about. These things have to be considered and treated if present. And then the first line treatment would be a bandage contact lens. So if a patient showed up in my office a week after PK and they had a non-healing epi defect like this, that looks pretty typical, I put a bandage contact lens, keep them on their antibiotic, maybe decrease the steroid just a smidge, and see them back three to four days later to see if there’s any improvement. If there’s no improvement, I would consider putting an AMT, and if that doesn’t work, I would do a permanent… Excuse me, temporary total tarsorrhaphy. You close the lid with silk suture. That generally can heal any kind of epithelial defect, unless the ocular surface is really bad, like unkeratinized. Then you shouldn’t be doing a PK anyway. But those are the only cases I’ve seen that can’t heal with a temporary total tarsorrhaphy. And if you want to cover with Valtrex, especially with a history of HSV, you would already be treating with that anyway, but you can treat with Valtrex if none of these things are working. We’ll talk a little bit about primary graft failure. So here’s a good example. This would be a patient — it generally says a month. That’s kind of the rule of thumb. If they don’t heal in a month, if they’re still cloudy like this, that’s primary graft failure. Why would a graft fail? Number one would be surgical trauma, the surgery was just too rough. But some elements of the endothelium, even if the pictures and specs look good on the tissue, sometimes the cells just — they don’t wake up. They don’t want to start working. And you get this cloudy cornea. This happens pretty rarely here in the US. Maybe about one every two to three years I’ll have a patient like this. It’s way more common with endothelial transplants, but that’s a much rougher surgery on the transplanted tissue. So you have to regraft. That’s the only choice in this scenario. So it’s always good to keep track of the size of the graft that you use, so you can use the same one. And when you do this surgery, you just go and you cut the sutures and you can basically pull out the old graft without having to… You don’t retrephinate, obviously, when you do this surgery. The graft hasn’t really scarred in nicely. You just cut the sutures and pull out the graft. Other bad things that can happen. Keratitis. If you’re gonna get an infection, a lot of times there’s a loose suture. You put fluorescein in the eye and you can see the outline of the suture. If there’s an infection, you want to see — is it on the surface or down deep? You want to do a culture and you want to start them on antibiotics, obviously. These can be things like, you can start them on Polytrim and gentamicin or if it’s bad, fortified vancomycin. For vancomycin, 25 milligrams per milliliter and tobramycin, 20 milligrams per milliliter. There’s something called crystalline keratopathy, a suture-related infection, which happens with a not so virulent form of strep viridans, and that can cause kind of a crystalline pattern, and that of course needs to be treated with antibiotics as well. Fortunately it’s very rare to have a fungal suture-related infection, but if you were treating them for fungal keratitis with a therapeutic PK, and they start with an infiltrate, you’ve got to assume that it’s fungal. And that’s kind of a topic for another talk, because fungal keratitis is a whole nother ball of wax. Fortunately pretty rare with just uncomplicated normal penetrating keratoplasty. We’ll talk a little bit about increased intraocular pressure. Glaucoma occurs in a fair share of PK patients. And it’s obviously more likely in patients who have preexisting glaucoma. Here’s the thing. You have to keep seeing your PK patients. You can’t just do the surgery and say bye-bye. They can get glaucoma years down the road. They get cataracts as well. And you have to keep checking the pressure. Glaucoma can lay dormant for years and spring up again and cause undetected blindness in your patients and they’re looking at you and saying… Hey, doc, why didn’t you see this? Why do they get glaucoma? They’re on steroids, for one, but you’re stretching the anterior segment quite a bit, especially in the immediate postoperative period with sutures, so you can have a change in the shape of the trabecular meshwork, causing glaucoma. Whenever I detect a high pressure, obviously we treat it and we taper the steroid, and if it’s left untreated, it can cause graft failure and cell loss. So anyway, you’ve got to pay attention. Always check the pressure on your PK patients. Okay. This is not, unfortunately, an uncommon complication. The patients are a couple of months out from surgery, and they say my eye hurts, I rubbed it, I felt something, my pupil looks funny, or the vision got really blurry. They come in and you see sutures missing or broken here, and you’ve got iris coming out of the wound. How do you deal with this? Obviously you’ve got to fix it. Fixing it involves, number one, kind of determining how long the iris has been out of the eye. I would try to save the iris. I wouldn’t be aggressive and cut it. I would remove the epithelium and then suture. If it’s been out for a week, you’re in trouble and have to make the decision about cutting it versus putting it back in the eye. I’ve had patients three to four days out, we can still put the iris back in, especially if it’s just a little bit. This can be done in the theater. If you’re messing with the iris, I would do it in the theater. You can give them Valium, topical anesthetic, reposit the iris, and do it in the clinic, but especially the extent you see in this photo, I would do it in the OR or the theater. Here’s another thing. Broken sutures without iris dehiscence. Or iris exposure. Expulsion. You have to suture these up. Again, there’s gonna be some sort of mild trauma that the patient reports. This could be two weeks out from surgery or two months out, three months out from surgery. You need to resuture this. Obviously if you need to reposit the iris, this could be done in a clinic in a minor procedure room with topical anesthetic, or it could be done in the OR with topical anesthetic, even. Okay? So if sutures are broken, you’ve got to fix them. This leads to why I like the running suture in addition to the interrupteds. All right. So we’ll talk a little bit about rejection. Obviously there’s different layers to the cornea, so there’s different types of rejection. This is epithelial rejection, which is pretty rare, and it’s the least scary of all of them. Okay? This is stromal rejection. Again, this isn’t as common either, but it kind of presents with these non-specific nummular opacities. And it’s treated with steroids as well. And here’s the one that everybody knows, and it’s on the test questions. It’s this endothelial rejection. You see the KP in a linear pattern, inferior edema where the inflammation has already occurred, named after Dr. Khodadoust, so obviously this graft isn’t gonna do so great. But you can reverse these, and the KP will always be left there, but in fewer numbers, and you can reverse it and get some more mileage out of this graft. So I treat these aggressively. So a little bit about the treatments. It’s pretty easy. It’s steroids. How are you gonna give them the steroid? You’ve got some different choices. Obviously topical. If you have Durezol, you can bomb them with Durezol six times a day. That’s gonna reverse most of it. If the patient is a little bit squirrelly and you don’t know if they’re gonna use the medication like you need, you can do a sub-Tenon’s steroid injection of Kenalog. The only problem is it sticks around for a long, long, long, long time. So if there are any pressure issues with the patient, you know the Kenalog can cause problems in the long term. So oral steroids are a little easier, I think, a little more controllable. You can put them on prednisone or Medrol dose pack. If you’re gonna treat them with prednisone, I do half a milligram per kilogram, something like that, or even 1 milligram per kilogram, just for a short period of time, a week or so, normally enough in conjunction with topical. If they’re not using their drops or you don’t know if they’ll take the medicine as indicated, sub-Tenon’s is good. If they’re diabetic, oral steroids can worsen their hyperglycemia, it can also have side effects such as… In the short term, it makes people anxious. It can cause mood issues and energy issues, and so some people you shouldn’t use the oral steroids, and you’ve got to find out if they’ve taken them before and how they’ve reacted to them. Okay. We can continue. So we got a question from Dr. Duong Mai in Vietnam. The question was: How to manage the graft for HSV and VZV. This is a beautiful picture I found on the internet. It should be in the Ophthalmic Photography Section. Obviously these are dendrites, but it isn’t a PK patient. I just wanted to include it because of the picture. If they have HSV or VZV, you have to decide: You’re doing this transplant or not. If you are, there’s got to be a good reason. Is the other eye affected? Not affected? A lot of people you would hold off on surgery and let them use their good eye because of the high risk of recurrence and graft failure. The success rate in corneal transplant overall is 80% to 90% if it’s keratoconus or scar, but with HSV, VZV, it’s less than 50%. If there’s a good reason to do it and the patient knows the risks involved, you want to treat with oral Valtrex in the perioperative period three times a day, and in the postoperative period, every day for life. You’ve got to watch for vessels, epi healing issues, and pressure. All more common in these patients. Basically managing a graft for HSV — you have to decide if you want to do it, if it’s gonna improve the patient’s quality of life, and secondly, you’ve got to cover them with Valtrex or acyclovir, valacyclovir or acyclovir for life, you have to see them more often, and you have to be on the watch for complications, which are more numerous with this type of problem. So that’s how you manage it. This is the question I get most from people. How do you manage the astigmatism after corneal transplant? So there’s different ways to do it. This is the way I do it. Dr. King in the United States asks: Do I prefer interrupted, running, or combined? I prefer combined for two reasons. You’re not gonna get a wound leak when you have 24 sutures in the eye, and I can take them out starting at 4 weeks instead of 6 to 12. Back in the day when PK was the only game in town, a lot of studies looked at the best suture pattern. So this was in the 70s, 80s, 90s, before we had DMEK, DSAEK, and everybody was doing PK. Basically they’re all the same once you’ve taken out the sutures. I repeat that. They’re all the same once the sutures have been removed. Whether a beautiful 25 bite running suture or crooked ugly 16 bite interrupted suture. Once they’ve all been removed, the corneas end up the same. And about ten years ago, when we started getting better femto lasers for LASIK, people started using these for transplants, and we had some studies. You could see this one in Cornea in 2016 showed that femto PK doesn’t do any better than normal PK. No less astigmatism. In early post-op, they look better, but once the sutures are out, it doesn’t matter. The bottom line is: Don’t pick a suture pattern because of the astigmatic characteristics of the suture pattern. Pick it because you’re good at it, you like it, and other reasons. That’s why I do combined. I can take out the sutures early, leave the running sutures in place, and less chance of a wound leak. So the way it works is I do a penetrating keratoplasty, you follow the patients, like we looked at in the early visits, and starting at 4 months, I bring the patients in. To take out the sutures, you need a topographer or the small keratoscopes, which is a little thing where you can project mires onto the cornea. Based on the topography, you do selective suture removal. My pattern is a 12 by running, 12 by interrupted. The interrupted sutures are fairly tight and they’re all the clock meridians. The running suture is not as tight and it is not full thickness. Just about 50% thickness on both sides. How do I know it’s not as tight? During the surgery, I pressurize the eye a little more when I’m putting in the running suture and it stays pressurized because it’s already got 12 interrupted, and since I’m putting it in at a higher pressure, it won’t be as tight. It’s when you suture the eye when the eye is soft and then you pressurize it that you have very tight sutures. Anyway, this is pretty typical for a patient of mine. 13 diopters of astigmatism and it’s right here at 150 and across. And I’ll take out one suture here and one suture there. I do that at the slit lamp. If you’re good, you always put the knots on the donor side, just so you have an easier time getting the sutures out. What you don’t want to do is take out a suture and rotate it through the interface, because you can cause a gape in the interface and a leak. Okay? So let’s say the knot is right here. Then I cut the suture here. And then I pull towards the center. Oh, sorry. Back. You pull towards the center. And you remove the suture. Okay? And then you put in an antibiotic drop afterwards, and before, and then you have the patient come back a month later. So, again, starting at 4 months, I bring the patient in. You measure the astigmatism. You don’t have to refract them. Just do topography, or use the keratoscope. You find the steep axis and you remove interrupted sutures, 1 or 2. Okay? If it’s all on one side, you would just take out maybe two on this side. I try to take out two sutures a visit. Okay. Then they come back 4 to 6 weeks later. You do another topography. Now the patient is down to 7 diopters of cyl and you use the topography and find the corresponding axis and you want to see if there are some sutures there. If there are interrupted sutures there, I take one out here and one out there. Again, antibiotics before and after. And then you bring them back in a month, and wow! Look. They’re better. 3 diopters of astigmatism. I can quit at 3. You can see the mires look more regular. Also, when else do you quit? So you quit when you have 2 to 3 diopters of cyl, or the axis of astigmatism — you don’t have any sutures that you can take out. So I’ve already taken them out over here. If the patient has 3 diopters, there’s not much I can do to decrease that further. If there’s no interrupted sutures to remove. Okay? Now… What happens if you take out all the interrupted sutures or at least all of them that are in the steep axis, and they still have high astigmatism? Well, in that case, you have to take out the running also. At least with my pattern. And so then I would wait. I would say… Okay. You’re kind of stuck here for a while. We’ve got to give it 8 to 12 months to heal completely, and then I’m gonna take out the running suture, and then I think you’re safe without a suture in, and then we’ll talk about doing another procedure. And those procedures would be astigmatic keratotomies at the graft-host junction, or if that doesn’t work, a wedge resection. Now, astigmatic keratotomies are done on the steep axis as well, and you want to do 20 degrees on either side, and you have to use intraoperative pachymetry to measure and you use a dynamic cutting knife and cut 80% thickness. It’s kind of an art. You can find some articles on it, and it’s an unpredictable type thing. But that’s what you do when all the sutures are out and they still have high cyl. Again, we’re gonna run through that, just because it’s the most critical part of the talk. I didn’t practice what I preach and you can see some of my sutures are still on the graft side. I’ve been more diligent about this in the last few years, but this is from five years ago. So you bring them in, you do topography, it’s still steep at 90, I take this guy out and this guy out and you bring them back in a month. What if I take out these three top ones, these three bottom ones, and they still have 6 diopters of cyl? That means you have to wait a while, and you have to take out the interrupted and you have to wait about 12 months for that. You can’t do it in the short term. If you take out the sutures in the steep axis and they still have high cyl, you have to wait for the cornea to scar in and heal a little better and come back and do some sort of ancillary procedure to get it fixed. The first choice would be astigmatic keratotomy and the second would be a wedge resection. Also you can do PRK. That’s another option. It could be more expensive. You want to make sure the cornea is fairly regular to do PRK, but that’s an option as well. Okay. We’ll move on a little bit. We’re gonna talk about two aspects of intraoperative surgery that you can kind of pay attention to, to avoid post-op complications. These are related to wound leak, epithelial healing, and astigmatism management. So you’ve got to avoid improper trephination like this guy’s hair cut. You’ve only got one chance. And you have to make sure that you always cut the donor first. Okay? So that’s… Even an experienced surgeon should always cut the donor. You should always confirm the trephine sizes with the nurse, to make sure your donor is larger than your recipient. Cut the donor first and again confirm. If the donor is too small, you have to suture too tight, you’re gonna get hyperopia, glaucoma, or wound leak. If it’s too big, the I looks ugly, and you have complications. Always confirm the trephine sizes, always cut the donor first. I oversize the donor by 0.25 over the recipient. Not everybody has these trephines available, but they’re getting cheaper and cheaper. I love vacuum trephines, especially for the recipient. For the donor it doesn’t need to be vacuum, but I would cut it in a silicone or trephine block like this. I wouldn’t just cut it on the stand, on a flat. Because the contour isn’t right and the cut is gonna be bigger. So again, vacuum for the recipient is fantastic. I highly recommend it. And for the donor, it doesn’t have to be vacuum, but it needs to be on a teflon block. Here’s a little video showing donor preparation. This is a typical teflon block. We put the donor in there. You notice I marked with a purple marker this little groove in the block. The reason I do that is because it helps me to center the cornea better. If I know where that purple ring is, I can get equal representation around the cornea, especially if there’s arcus, to make sure that it’s centered nicely. So I keep moving it around, not touching the endothelium, obviously. Until it’s nice and centered. So now I see the same amount of purple all the way around. This is a bigger graft, and then I punch it. Here’s another trick. You rotate it on its side and spin the corneal-scleral rim to make sure it’s cut 360, you lift it up, and you’ve got a nice donor right there. The other thing is you want to avoid misalignment. We talked about suture. You want to be 90% thickness like this. You don’t want to have a 50% thickness. You don’t want to be 100% thick. You’re gonna get wound leak and epithelial migration. After you put the four cardinal sutures in, you want to dry the corneal surface with a Weck-Cel and you want to see this diamond. That tells you that you have equal pressure and equal distribution of the corneal tissue. If you don’t get a diamond like this, it’s more like maybe pull to one area, one suture is too tight or you don’t have the graft centered nicely. So… Obviously we know the first and second sutures are the most important. But also we want to… To make sure we mark the cornea nicely. So you measure the corneal diameter there. All this is helping you to get a good graft. And you cut… You calculate half of it, and you use the calipers to mark the geometric center of the cornea. Not the pupillary center. The geometric center. And I mark with a Sinskey like that, and you measure and confirm that you’re good. When you use the vacuum trephines, and here’s another way you can do it. You use a smaller diameter trephine and you center it, so you have some marks inside of your trephine that you can see. But the vacuum trephines have a little cross hair that you can use to aim the trephine. Also here we’re using an RK marker. That helps you, especially when you’re starting out, to know where to place your sutures. So all of those things are helpful. Those markers are very cheap. You can buy them at almost any conference. All right? And then we’ve got those marks. We look at the cornea again. Use some 0.12s to center it, and I’m looking for my mark that I have, and I’m looking for any dots that I have, all in the inside there, and you have an assistant who puts the trephine on suction, and then you begin to trephinate. You go about 75 microns each quarter turn, and some people like to enter the AC. Others don’t. It kind of depends on the patient for me. If it’s a pretty regular cornea, you can be sure you can enter it pretty safely, as long as your assistant is reliable and they come off of the vacuum. Here we’ve got a little aqueous, we removed it, and we find the area that was open, and I always like to be there at 11:00. This is the patient’s foot side here and this is the head up here. And I use a cohesive viscoelastic, so it can be evacuated easily, and you go around with the cornea scleral scissors. You want to peel it. You want the scissors perpendicular once you’re cutting with them, so you don’t have a ledge — a shelf, rather. You want to be real thorough. You always want to visualize the inferior side of the scissors. These are nice curvy ones. Some like them less curvy. I like them curvy. And they’re blunt, which is good too. You don’t want to go poking in the iris. And then we go all the way around. And remove that donor. Go ahead and advance. Then of course putting the two sutures in. This is the critical aspect. So we’ll start the video here. So I’m adding a little more cohesive viscoelastic. There’s some hemorrhage, but it’ll stop on its own. We bring the donor over with a spatula and lay it in its bed. Want to make sure it centers nicely there, so we’re good. And I have the marks. So I know that’s 12:00 right there. I did a 90% thickness using the double edged forceps. These are Colibri forceps. And you put the needle underneath where your forceps are, and you want to come out just short of the limbus and push through nicely. And then in terms of tightness, you want them fairly tight. I lock the sutures after I put the first three knots. So we do a 311 surgeon’s knot. I always lock it back onto the donor like that. So it has to be pretty tight. And I was just removing some blood from that area. You always want to grab the tissue with the forceps and make sure it’s equally distributed. This is easy because we know the marks are there and they’re good. It’s like painting by numbers here. You pass, give yourself an assist here, and then you have — coming out just shy of the limbus. Once you have those two in, you tie them. Obviously we’re gonna do the cardinals next and look at that. Okay. I’m just gonna rehash a little bit about the post-op astigmatism management, and then I’ll take any questions from the audience. So we bring the patient back in 4 to 6 months, they have 12 interrupted, 12-by running suture. We look for the steep axis and remove the sutures on the steep axis. About 2 on each visit, use antibiotics before and afterwards, and bring them back a month later. Less astigmatism. We find sutures in the steep axis and we take, again, two out. They come back in a month, they only have 2 to 3 diopters of astigmatism, we stop. If they still have high astigmatism, let’s say it has 6 and it’s still in this direction, you’re kind of out of sutures to remove, because you’ve already taken them out here. You have to wait for the cornea to heal more, and then you can remove the running suture and see what happens to the shape, and if it fixes it, you can pat yourself on the back. If not, you have to talk about doing a procedure, which could be something like PRK, it could be astigmatic keratotomy is what I like, because it’s cheaper for the patient, and to do PRK, I really would wait until all the sutures are out, so you’re waiting a long time. So our objectives in our lecture would be to be able to recognize and manage post-PK complications, so we ran through all of those, including the most common, which is an epithelial defect postoperatively. We also learned how to prevent these by doing good intraoperative management. So we’re gonna confirm trephine sizes and make sure we have good tension on the sutures. And then we’re gonna know how to recognize and treat allograft rejection. We know what it looks like in a slit lamp. The patient will come to you and say: I have a red eye, I’m sensitive to light, and the vision is foggy. I tell all my patients, if you get those three symptoms, you need to come in and let me look because you can have rejection. But people say — I’m always sensitive to light. You have to say: I’m so sensitive I can’t go outside and it starts tearing and it hurts pretty bad. Severe light sensitivity. Lawrence, if you can pull up these questions… Which is a medicine we don’t use to treat graft rejection? Everyone should get this right. You can show the result there, Lawrence. All right. We’re good there. Just a handful of people weren’t listening. That’s okay. Maybe they got in late. Which of the following do we not use to treat a persistent epithelial defect? Which is not a modality? The answer is Latanoprost drops. I don’t like to use it because it has a risk of CME and a risk of increasing inflammation. After successful PK surgery in an adult, one should begin removing sutures in one month. True or false? We’ll pull that up. Show those results. We don’t want to start taking out sutures that soon. So I’m gonna… Show you this video. This is the end result, of course, again. I want to thank you for your attention, and I’ll answer some questions now. We got a question who said: What do you do with a steroid in case of HSV epithelial keratitis on the graft? In a normal patient, you would stop the steroid and use… Okay. This is about treating HSV. The way that it works… If they have an epithelial defect, you want to treat them with Zirgan or acyclovir. You want to stop the steroid, because it can make it worse. That’s the same case with PK. Once the epithelium heals, you can safely treat with steroid. Especially for stromal HSV. But let’s say they just have a dendrite alone. I would stop the topical steroid. I would put them on Zirgan, or some sort of topical acyclovir or ganciclovir, and maybe switch them to an oral steroid for a period of time so you don’t get problems with the graft. If they had a graft for a long time and you can safely treat just with Zirgan, that would be okay, but if there’s an active dendrite, you should stop the topical steroid. So I answered that one. The next one. Anonymous attendee. How do you differentiate endothelial rejection from herpetic endotheliitis? That’s a good question. Herpetic endotheliitis can affect the good eye, but it can also affect the graft. Most endothelial rejection is gonna present with that line, and it’s not gonna affect the graft. Herpetic endotheliitis has clustered KP with diffuse edema. Whereas endothelial rejection has sectoral edema involving the Khodadoust lines and the KP. A little bit different. If somebody has preexisting herpetic disease, if they present with rejection type KP, you’re gonna give them Valtrex anyway, okay? But for somebody to not have any kind of history of herpetic disease, if they present with KP, and graft edema, I’m gonna assume it’s allograft rejection and not herpetic. But let’s say they don’t respond to topical steroids. You could add Valtrex in that same situation. To summarize that, because it’s complicated, they look a little different. The allograft rejection is Khodadoust line with edema that’s sectoral. Herpetic endotheliitis is clustered KP with diffuse corneal edema. If a patient has a history of HSV and present with any kind of rejection, you’re gonna treat with increased Valtrex and increased steroid. If they don’t have any HSV history and you assume it’s allograft rejection, but they don’t respond to the steroid treatment, you add Valtrex. How do you measure IOP post-PK? We talked about this. The best is gonna be a Tonopen or pneumatonometer or Goldmann if you can do it. Role of PK trabeculectomy to control IOP? This is Dr. Sujit Biswas. You don’t want to do surgery after PK. The PK is gonna be the last surgery you do. You’ve got to fix the glaucoma first. Otherwise you run the risk of jeopardizing the health of the graft. So if they develop post-PK glaucoma, you don’t have much of a choice. But you try to take care of this before they need the PK. And you can do a trab and then do a PK. It’s fine. But you’ve got to get the pressure — you’ve got to get the pressure fixed before the PK. Dr. Biswas also asks: Is there an increased risk of rejection after tarsorrhaphy? No. However, if you do a permanent tarsorrhaphy, it’s hard to see the cornea to make sure it’s okay. So you want to leave enough room medially where you can pry the eye open at the slit lamp, have the patient look medially, and just investigate the graft, and make sure it’s okay. Which is better for PKP? A continuous or interrupted stitch? In my opinion, a combined is the best. Continuous alone is technically difficult and it’s difficult to adjust any astigmatism post-op. You’re pretty much stuck with what you’ve got. Interrupted sutures, number one, they don’t look pretty. Number two, if one breaks, you get a big old gap, and you’re gonna have a wound leak. Dr. Awe asks: What should be done if there’s less than 2 diopters of astigmatism before suture removal commences? If it’s good from the beginning? That’s a good question. The answer to that is nothing. You just leave all the sutures in place. And you just monitor for any vessel growth or loose sutures over time. You just give them glasses and not take out any sutures. If one breaks, you remove it, follow them, let them know what a broken suture feels like, in case they have to come in. But if they’re good and you don’t have to take out any sutures, good work, doctor. Dr. OPD asks: How do you know if the donor endothelium is good for graft? This is more of a question for preoperative planning. Obviously you want endothelial cell counts and you want to look at the picture. So in many countries, they might have the endothelial cell count, but you could do… At a slit lamp, you can do cell images like Purkinje images, and look at the graft. If it was trauma, it’s probably gonna be good endothelium anyway, but most eye banks that I’ve seen in India, Peru, China, Myanmar, they have cell count. You want to look at the cell count, the picture, and the heterogeneity of the cell to make sure everything looks good. When is a suitable time for cataract surgery after PK? Six months is what I do. You want the astigmatism sorted out before you do the cataract. Because I’m not saying you’ve got to put in a toric lens or something, but you want to know kind of what the post-op refraction is, and it’s stable, and where they’re gonna be, and then you do the cataract surgery. I wait six months. That’s normally enough time. It gives you three chances to take out sutures and most of the time it’s a good time to do the surgery. On a side note, cataract surgery can be quite tricky after PK. If you’re not used to doing it, maybe do it with a block instead of topically, so you’ve got some time and the eye is not moving, because the visualization can be poor. Even if it looks clear at the slit lamp. Dr. Mamaclays is… Is it advisable to maintain on lifelong IOP-lowering meds? If they have high pressure, yes. You follow them like a glaucoma patient every six months. Most graft patients won’t require it, but if they’re on any topical medicines for IOP, you need to follow them like a glaucoma suspect patient. Every six months, alternate between OCT and Humphrey visual field. Dr. Ereca asks: How do you differentiate rejection from failure? Graft failure is gonna happen early or late, and rejection happens late. A graft that’s rejection — the conj is gonna be injected and you may have infiltrates or endothelial rejection line. A graft that fails gets edematous in a quiet eye. So if the eye is inflamed, it’s rejection. If the eye is quiet, so let’s imagine you did a graft six years ago on a patient, and they were doing well and they come in, and they say their vision is cloudy, you look at them, the pressure is okay, but the graft is edematous, maybe some stromal folds, you do a cell count but it’s not a great picture because the corneal is cloudy, that’s failure. If the same patient presents with photophobia, red eye, and it happened quicker, that’s gonna be rejection. In any case you can treat them with steroids and make them a little bit better and that gives them a little more time to consider regraft. In the absence of any suture-related complications or astigmatism, when do you remove the sutures? Again, if there’s no problems, I leave the sutures in forever. If the sutures have blood vessels grow into them or they break, I take them out. People can live with 10-0 nylon sutures in their eye without any problems. If the shape is good… If you start taking them out, you’re gonna mess up the shape and say… Doctor, I was doing so well. Why did you take out the sutures? Another question. You mentioned PRK to correct residual astigmatism. Is it not gonna induce significant inflammation and trigger rejection? It’s a tricky deal. You need the right patient with astigmatism. The best case is all the sutures removed, two to three years out from surgery, they’re already behaving well and the graft is doing well. You don’t want too much astigmatism. You want three or less, and they want it as good as it can get. It can damage the endothelium. Even though PRK has been shown to be healthy in healthy eyes, in PK eyes, you can cause graft failure by killing endothelial cells with that laser, so you want to do a cell count and make sure the graft is going great. It won’t really trigger much inflammation. Or rejection, I don’t think, but it can cause damage to the endothelial cells. The femto laser is banging on that cornea, and if the cornea has a 900 cell count, and they see well through it, I wouldn’t PRK a 900 cell count eye. A 2,000 cell count eye, I would feel more comfortable. Dr. Sandar asks: If a patient has inflammation, how many months should you wait for transplant? You wait until you get the inflammation fixed. The eye has to be quiet or else you’re gonna have an unhappy patient. When do you stop steroid medications, post-PK? We covered this in the talk. Never is the answer. So you want them on Lotemax or fluorometholone every day or prednisolone acetate every other day for life. Anonymous attendee: In case of suture abscess in the early postoperative period, would you remove the suture without risk of having a leak? And if you remove the suture, how would you treat if the leak happened? So if there’s a suture abscess, you’ve got to take out the suture, because it was loose. A loose suture is not doing anything anyway. So when you take it out, it’s not gonna cause an increased risk. It’s either gonna leak with it or leak without it. So again, if there’s a loose suture, you always take it out. If it’s loose, it’s not keeping anything tied together. You need to culture that suture. You put it on a plate like a chocolate agar plate or you put it in a thioglycolate broth. The suture is gonna help with the culture. So if you have a suture abscess, you remove the loose suture and you put them on fortified antibiotics and culture the suture. Anonymous attendee. Do you recommend phaco or SICS as preferred mode of cataract surgery, post PK? I recommend phaco. If the patient has a cataract so bad that you need to do small incision, like it’s a mature or very dense cataract, you need to do a PK triple at the time of the original surgery. What I’m talking about is cataracts that are mild and they worsen in the postoperative period. But if somebody has a bad cataract, you take it out open sky at the time of transplant. Anonymous attendee: Do you recommend 10-0 nylon suture or 9-0 nylon suture. 10-0. 9-0 is too big. Another one. Do you use topical cyclosporine or tacrolimus instead of steroid after tapering for long term maintenance therapy? I think the answer is no. If you’re using normal topical cyclosporine, the commercially available Restasis, it’s not strong enough. Now, if you get it compounded at a higher dosage, it can perhaps — I’ve seen some literature supporting it, used long-term, but normal Restasis? Not strong enough. Tacrolimus — if the patient has other diseases, like atopic dermatitis, et cetera, you can do tacrolimus, but it doesn’t replace steroid. All right. Another question. How do you remove a suture? Do you pull on the donor or host site? We went over this. What you want to do — it doesn’t really matter, as long as you don’t pull the knot through the interface. Okay? That’s what you want to avoid, because that’s when you can cause it to gape. Do you ever recommend RGP before complete removal of suture? Yes. So… No! No, no, no. You want to try to get the astigmatism down before you would do an RGP. So no, you would want to — an RGP would be required if a patient had high astigmatism after you removed all the sutures. Don’t just go to an RGP straight away. You want to make sure you try to get it down by removing the sutures first. And then lastly, when do you use oral steroid over topical one for graft rejection? I use them at the same time. If a patient presents with rejection, I would put them on topical and oral steroids. The topical one would be… I mean, the steroid rejection would be… If you use steroid with rejection, it would be in combination with the topical one. I think if it’s bad, you use the oral steroid. If it’s mild, just use the topical. Okay, another question. I’ve got to go here soon, folks, so I’ve got to cut off the questions here in a little bit, but I’ve got four more I’ll answer. When do you plan other eye PK in case of bilateral disease? I would wait six months to a year, in a PK. If it’s a DSAEK or DMEK, a month. But… No, DMEK, a month. DSAEK, three months. But a PK, you want to make sure they’re doing well with it, before you move to the other eye. Another question. Do you recommend using a hard contact lens with suture still in the graft? We just answered this one. No is the answer. You want to try to get the cyl down removing the sutures first. In case of nummular stromal opacity, how do you differentiate between infectious keratitis and rejection? It’s gonna be rare that you get infectious stromal opacity alone, unless it was that crystalline keratopathy, and that will start at the suture, because it’s gonna come in from somewhere. The nummular opacity — they just occur sporadically at the center of the graft. That’s very unlikely to be infectious, unless there was an epi defect where the bugs got in. Now, if the patient had preexisting fungal keratitis or something like that, that you did a therapeutic graft for, and then you see an opacity, it could be fungal. Fungus is different. But with bacterial, 9 times out of 10 it’s gonna be a rejection, the stromal opacity. Also, it takes a long time to get stromal rejection. So if this happens in the postoperative period, right after the surgery, it’s gonna be infectious. If it’s years later, it’s gonna be rejection. And then last question. When do you use IV methylprednisolone? I would rarely use it. I don’t think it’s totally indicated for graft rejection. I would use that more in patients that you… I guess you’re treating optic neuritis and stuff like that. I don’t really see a role for that, unless… I would do depot Kenalog and oral steroids. Unless you had to admit the patient because they were from out of town and you needed a good reason to get them in the hospital. Something like that would work. All right, guys. I want to thank you for your attention. It’s 9:00, so I need to go to clinic. I’m glad I got so many good questions. We’ll see you next time. Thank you very much.

Download Slides


November 25, 2019

Last Updated: October 31, 2022

Leave a Comment