Lecture: Problem-Solving DALK

[Nashrah] As I mentioned, my name is Nashrah and I work at a non-profit organization called SightLife. SightLife has a worldwide mission to eliminate corneal blindness. And I just wanted to share with you a little bit about how we do that across all the geographies that we work in. So SightLife has five key programs in which we function in. Advocacy and policy, prevention and awareness, clinical training, eye bank development, and innovation. Through these five key programs, we’re able to help countries transform their own health systems, so that they can be self-sustaining, to really prevent blindness and restore sight to those in their own communities, in their own areas. This is critical for us to see people, over time, come and get the access they need for care. By closing the gaps, the patient can be centered as the key recipient or beneficiary, if you will, from the health system changing.
So through this we’re able to really see a shift in the corneal blindness community. And with the goal, ultimately, that nobody has to wait for a cornea transplant and any corneal ailment that could be affecting a patient can be addressed sooner than later. We are ultimately a resource. For corneal surgeons, worldwide, we want to be a resource by providing the proper education and access to tools that they may need. Specifically the clinical training program, which is the program that I’m a part of, we focus on getting education to the proper individuals in the ophthalmic realm. So this covers ophthalmic surgeons, general ophthalmologists, allied ophthalmic personnel, optometrists. Anyone who could potentially help a patient with a cornea problem, we want to make sure that they have the ability and the knowledge necessary.
And we do this in a number of ways. So as I introduce SightLife to you, I want you to be aware of the innovative opportunities that we have ultimately through our partners, like Orbis and Cybersight. This curriculum that we built is a peer-reviewed curriculum that is competency-based from experts all around the world. Engaging with faculty, ultimately, is the best way for us to get across critical pieces of education to ophthalmic professionals, so they can serve as coaches and mentors for hundreds of corneal surgeons worldwide. And ultimately this is the cohort of individuals that helps us create an environment of learning, non-stop in your career, so that your surgical skill sets can continue to grow over time.
It is my ultimate pleasure to get to work with someone like Dr. Roberto Pineda and I’ve been lucky to work with him for a number of years. So I’m hoping that everyone really enjoys this webinar today. I do have my contact information on the screen, if you don’t have it. Please make sure to write down my email or my phone number and I’m available at any point in time should you need access to additional mentors. But without further ado, I’d love to pass it over to Dr. Pineda.
[Roberto] Thank you, Nashrah. And welcome everyone, as-salamu alaykum. Thank you for being here at the end of Ramadan and the start of Eid tomorrow. I have the pleasure of talking today about one of my favorite, yet frustrating, corneal procedures, deep anterior lamellar keratoplasty. And talking about some of the problem solving issues around that. And also giving you an overview of this very satisfying procedure when it works well.
I have some questions that have been sent in, but obviously we’ll talk about more of these. I have a lot of slides here and some animations, as well as some videos to talk about. And hopefully you can take away something very useful here.
There are my financial disclosures, none of which are relevant to today. I do want to acknowledge Dr. Gangadar for sharing some of his videos, as well as some slides from Dr. Anthony Aldave.
There’s been increasing interest in corneal lamellar surgery for selective replacement for corneal pathology over the past few decades. Traditionally, we have used penetrating keratoplasty as our choice for corneal stromal disease. But as we all know, corneal transplantation is fraught with complications. From graft rejection, to irregular astigmatism, to corneal opacification. And for those of you who have access to potentially posterior lamellar tissue, endothelial keratoplasty has actually become a very common procedure for endothelial dysfunction such as bullous keratopathy and Fuchs’ corneal dystrophy. And recently in the U.S. over the past few years it’s actually become more common than penetrating keratoplasty.
However, anterior lamellar keratoplasty, such as DALK, has only really increased about three fold since 2005. You can see here in the U.S., we were doing about 869 anterior lamellar keratoplasties, and in 2016 that number had only risen to about 2,400. So very slow increase relative to endothelial keratoplasty, which has probably increased by over 20 times.
But there is a lot of, as we all know, and I saw some of the participants had actually attempted or performed DALK in the past, we know there’s many barriers to this procedure. It has technical challenges including unpredictability for even seasoned corneal transplant surgeons. The surgical time is much longer than a standard penetrating keratoplasty. I can probably do a corneal transplant in about 30 minutes. DALK can easily exceed an hour or even an hour and a half. And in the U.S., we actually get paid less for this procedure.
As I mentioned before, we’re going to do an overview of DALK and talk a little bit about some of the literature. This is from the most recent Eye Bank Association of America, data from 2016. And you can see the total number of grafts performed in the U.S. has increased a lot over the past several years. And we’re doing around 83,000 transplants in the U.S. Penetrating keratoplasty, as I mentioned before, has actually been decreasing. It’s actually under 30,000 now. And anterior lamellar keratoplasty has only risen a little bit compared to endothelial keratoplasty, which is now eclipsed standard penetrating keratoplasty.
So what’s involved in the DALK? Just a quick review. We’re actually removing the corneal stroma down, ideally to Descemet’s membrane. And that means that this procedure is going to be most useful in corneal diseases for people who have normal functioning endothelium. And fortunately there’s been many advances in this area which have started to repopularize, I should say, this type of surgery.
DALK basically can be applied ideally to any anterior corneal pathology where there’s healthy endothelium. And probably the most common indication in the world for DALK is karatoconus. These are usually healthy patients, young patients with healthy endothelium. They can, in theory, have corneal scarring, and also other conditions where Descemet’s membrane is sparedm with corneal scarring conditions like herpes zoster or herpes simplex. You can apply to other ectatic corneal disorders such as pellucid marginal degeneration, as well as post-refractive ectasia.
Other less common indications for DALK include certain dystrophies, with the exception of maybe macular dystrophy, VKC, even indications where there’s ocular surface disease such as limbal stem cell conditions. And then metabolic storage diseases such as mucopolysaccharidosis, after active corneal ulcers, and even for tectonic indications.
Obviously that means there’s going to be some contraindications and those includes those where the endothelium is going to be compromised. And since we are preserving that portion with the DALK procedure, if the endothelial cell count is low, we’re not going to want to perform DALK. So that means even deep scars over the visual axis where there might be pre-existing defects or breaks in Descemet’s membrane, those might be relative contraindications. As we all know, that many keratoconus patients develop hydrops and in those cases you can decide if the tear is below the visual axis, you might want to consider a manual DALK as opposed to a penetrating keratoplasty.
Let’s talk a little bit about the history, I always think it’s important to know where procedures come from. DALK, the concept of DALK is not new. People were trying to do this basically 60 years ago. And it’s only very recently in the past 25 years that we’ve talked more about the actually DALK procedure. And this has included Anwar back in 1974. But it really wasn’t until Dr. Anwar’s paper in 2002, which described the big bubble technique, that this procedure has regained popularity.
What’s involved? If you’re going to consider doing DALK in a patient, what kind of information do you need before you make that decision or make the best decision for the patient? So the preoperative evaluation is very important for these patients. And the pachymetry, I think, is one thing that you absolutely need. You need to know how thin the cornea is, you can obtain that through ultrasonic measurements, you can obtain it through topography, such as a Pentacam, Galilei, Orbscan, pachymetry maps that can tell you how thick the cornea is. And that will help you in how far you have to trephinate as well. So that information is very important.
If you have access to anterior segment OCT, I think that’s another piece. It’s not mandatory but it’s very helpful, especially when you have scarring. You want to see how deep the scar is. And to get a sense of what the endothelium might be, the status of the endothelium, if you can’t visualize it well. If you have access to anterior segment OCT, which many people do these days, it’s a very helpful piece of information.
Ideally you want to verify that the endothelial cell count is healthy and we ideally do this by specular microscopy. Now we understand that many people don’t have access to specular microscopy, it is an advanced in-office device. But you need to do specular reflection. How do we do specular reflection? We can use scluraotic scatter, with the slit lamp, you can look at their red reflex and really focus on the endothelium to make sure that it’s healthy, and that the patient’s going to be an appropriate candidate for DALK surgery. But if you have access to specular microscopy you should obtain that for the patient.
And then in cases where there’s maybe a lot of edema and you can’t really see the endothelium that well, again, if you have access to confocal microscopy that’s nice. Again, this is usually reserved for very advanced tertiary centers, as most people don’t have access to this. But it is nice to have.
I have a few questions scattered throughout this webinar, just to get people thinking about things. But the first question is, which condition is least likely to benefit from DALK? One, keratoconus with anterior corneal scarring. Number two, keratoconus with this history of corneal hydrops. Number three, anterior stromal scarring after corneal ulcer. Number four, stromal scarring after HSV keratouveitis. And number four, corneal clouding from mucopolysaccharidosis. We hope this will get you thinking about what types of patients might be good for DALK and those that would not be considered reasonable candidates for DALK.
The condition the least likely? We have answers for B and E. So patients who would least likely benefit from DALK. Again, we want to look at those patients that either have compromised endothelium, or compromised Descemet’s, or compromised endothelium. So patients with a history of anterior corneal scarring would be a good candidate because their endothelium would probably be healthy. Patients with a history of corneal hydrops, this would really depend on where their hydrops occur, but you could do manual DALK for a patient like this, but not big bubble DALK. Anterior corneal scarring after a corneal ulcer would be a good candidate for big bubble DALK. And corneal clouding from mucopolysaccharidosis is actually a good candidate. I actually have a large series of these patients that I’ve been treating for over 10 years. And they’ve all done very well.
Stromal scarring with HSV keratouveitis. Now with keratouveitis you have an inflammatory anterior segment reaction. And that iritis or uveitis actually affects endothelial cell count. So these patients actually usually have a very low endothelial cell count. So this would be the patients that would be least likely to benefit from DALK. So if a patient with HSV keratouveitis, that is not the best patient for DALK.
Let’s go on. So these are just some case studies, I’m going to go over a few of these kind of quickly. Just to demonstrate who we think about would be a good candidate. So this 22-year-old man with decreased vision, has had frequent changes in his refraction. He’s not satisfied with his glasses and he’s been tried with contact lenses, but not able to tolerate them. And he’s referred to you for corneal transplantation.
You take a look at this patient and you can see they have relatively advanced keratoconus with some anterior corneal scarring. You look at topography, you can see there’s classic inferior steepening on topography on the right and a Fleischer ring here on the left. And then specular microscopy shows excellent cell count with the cell density of 2451. So this patient would be a great DALK patient. So this is probably a lot of the patient’s you’re going to see, I would imagine, that you would consider for DALK. And this is probably the most common scenario.
So here’s another one. Patient with keratoconus who had INTACS previously. We used INTACS to help recenter the cone and give the patient better directed visual acuity either with glasses or contacts. This patient has lost their ability to have acceptable correction with glasses and is referred for corneal transplantation. INTACS are intrastromal corneal rings that are usually placed in about two-thirds the depth of the cornea, as you can see here, in the kind of mid-peripheral cornea. And you can see the depth here is about two-thirds to three-quarters depth.
Cell count shows excellent cell count of over 3,000 for this patient. And so this patient would also be a DALK patient. You can take out these rings, actually, fairly easily. You can just cut down in front of the ring and there’s a little hole, I’ll show quickly here, right to the right of the slit beam. You can see there’s a little hole that you can grab and wiggle that segment out, and then you go ahead and do your DALK procedure like you normally would. I’ve done a few of these cases and they work quite well.
Here’s another patient. Now, many of our patients these days are unfortunately getting cross-linked to further keratoconus to keep it from progressing. This person was cross-linked but they had progression of their keratoconus. And we do see a failure rate of around 5%, maybe three to 5%, of cross-linked patients. It really depends on the treatment protocol that they receive. Bresden protocol is around that and it’s higher for children. So they were referred for transplantation. And we know that the corneal lamellar’s more compact. So that actually might be a barrier for the big bubble formation. But in this series of nine eyes they were actually shown they were able to successfully complete the big bubble in eight out of nine cases.
Confocal or specular showed no change in the endothelium cell count. So these patients can have DALK and this patient had a good cell count as well.
And then I think this is the last one, this is 25-year-old with a cloudy cornea. They had a family history of decreased vision. You can see here, I know that some of you may recognize this, have a lot of this condition in your area. This is macular dystrophy. So interestingly, macular dystrophy is one of these conditions where you think DALK would be potentially a good option. But this study actually showed that the recurrence, we know that macular dystrophy will come back in the transplant. But this one actually showed that the recurrence rate was about 17.5% in penetrating keratoplasty procedures, but nearly 43% in DALK. So the recurrence rate was five times higher in DALK versus PKP. So although macular seems like a really good indication, it may not be the best choice for these patients based on this paper. We would actually recommend if you’re starting, especially if you don’t have a lot of DALK experience, we would probably recommend PKP. But we’ll talk about how this type of patient might be good for practicing DALK.
This is the last one a 21-year-old with a corneal scar. Has multiple superficial keratectomy and PTK, or phototherapy keratectomy, with mitomycin. Sorry, that was an extra one.
I’m sorry, this is a 46-year-old with a corneal opacification. She had the herpetic keratouveitis two years ago, this is the appearance. So you think, okay, anterior corneal scar, we can maybe do DALK? But you see the specular microscopy is not very good, even with confocal microscopy here, she has the cell count of less than 1,000. And PKP would be the choice for a patient like this.
So let’s go to our second poll question here. This is asking which preoperative test is least likely to be required for DALK surgery? Okay, good, so we got some testing, everyone agreed that corneal topography is a good idea, we want that pachymetric map, if we can. Ultrasonic pachymetry we said is very important. Corneal sensation is important for grafts, in general. If we don’t have grafts, if we don’t have good sensation it’s hard to get re-epithelialization. Schirmer’s testing we do for dry, and specular microscopy we said patients need good endothelial cell count. But in terms of Schirmer testing. If you remember, we talked about how we can do DALK for ocular surface disease, even patients with mild Stevens-Johnson and OCP, ocular cicatricial pemphigoid. Those are dry eye conditions. So a dry eye is, we don’t have a requirement, but the patients can have dry and still have DALK. So of all of these, probably Schirmer testing is the least important to determining whether a patient can have DALK or not.
So let’s talk about some of the instruments we use in DALK surgery. In my mind, this procedure is a good procedure to learn because it doesn’t require a lot of special instrumentation. These are from Storz, they are the fogla dissector and cannula. And to be honest with you, most of you already have all the instruments you need to perform this procedure. Probably with the exception of these two instruments.
So the dissector, as you can see here, it’s actually sharp but kind of blunted at the tip. So you can make the dissection down deep into the corneal stroma. And then you’ll notice the cannula up above. It’s a 27 gauge cannula and you’ll see that the hole is actually on the bottom of the cannula. It’s designed this way so that you’re forcing the air down against Descemet’s to help separate it. And some people use a needle to create the bubble but the issue with the needle is it’s very sharp and the air is directed more horizontally than vertically towards Descemet’s. Now, you can use a needle and that’s been used very successfully. But I would say as a beginner when I was learning DALK, probably one of the biggest changes I made to help me perform DALK more consistently was to use a cannula. So this is an investment that I think is very, very worthwhile if you want to consider doing DALK moving forward. And so these two instruments, I think are key.
Now, scissors, there’s specialized scissors for this as well as spatulas for removing the anterior corneal stroma and dissecting the lamellar, the posterior stroma lamellae if you need to. But honestly I just use either a cyclodialysis spatula or my regular Westcott scissors to remove that. So you don’t need to spend money on this, but they’re nice. If you’re going to get a set, these are blunter and they have kind of asymmetric blades that allow you to remove it a little bit more safely. These are by Moria, they also make a set of instruments as well with dissector and cannula. And the Sarnicula has one and Fogla have one, which are probably the two most popular. If you have anything like a Drysdale, you can use a little paddle to help separate that, put that lamellar as well, that’s also a relatively inexpensive instrument that you can use for this technique.
Now, most of us, let’s talk a little bit about trephines. This is, for DALK, as you know, we try to trephine, again, about two-thirds to three-quarters of the way into the stroma. And this is, of course, in the beginning very hard to judge. You need, I would say, a very sharp trephine. You wouldn’t want to use one, I would not recommend doing one manually because the trephination’s going to be very, very uneven. So you want a suction trephine and I think you have to know how much the trephine is cutting per quarter of rotation. Most of the trephines cut anywhere from 50 to 100 microns in a quarter turn. But some of them, like this Moria, that’s shown here on the slide, you can actually pre-set it to a depth. Now, you pay more for this, of course it’s a disposable trephine. But if you can reuse it a few times for a patient, with alcohol in between. And this is ideal for the beginner because it’s going to prevent you from potentially perforating during your initial DALK surgery. But for a beginner, I think it is useful. I actually don’t use this pre-set trephine, but I do use a vacuum trephine for my DALK surgeries.
The other thing I want to mention is just the type of tissue. And one of the benefits of DALK, as you know, is you don’t need to have the healthiest tissue. And you might be able to use one corneal tissue for two patients. But this is one of the areas, if you have access to an eye bank, you can potentially use a less quality tissue. But you also might need a backup tissue in case you do perforate and you need to do a penetrating keratoplasty.
In the beginning, you probably need to have access to two tissues. One that’s not as good that you could use for the DALK and the other one that you can potentially use for regular penetrating keratoplasty if needed. We’ll talk about that a little bit more. Obviously you have to consider the risk of perforation, some conditions are much more prone to inversion and you might need to schedule a DALK one day with a PK for backup next.
Let’s talk about the technique, most of you are familiar with the DALK technique. I do think everyone has little variations. But I do think there’s some key points to make sure that you successfully perform DALK. I always mark the geometric center of the cornea and then mark the cornea with an RK marker of some kind as shown in A. I usually, as I mentioned before, set the corneal trephination depth to about two-thirds depth. In the U.S. and some of the western countries we have access to femtosecond laser. So people have talked about femtosecond DALK. That’s a very obviously expensive option but you can very precisely determine the depth of the oblation. And if you were a beginner, this would be a good experience. But I think getting experience with the trephine is going to be the way more people do their DALKs these days.
And then depending on the condition, you’re going to potentially remove the anterior corneal stroma, particularly if it’s opacified. So you can see your placement of your dissector and cannula or needle. Or if you’re going to be performing a manual DALK. Then once you’ve created, used the dissector and the cannula, we’re going to go ahead and inject air. Create the big bubble, if we’re doing a big bubble DALK, as opposed to a manual DALK. And again, you want to get ideally, within about less than 100 microns above the Descemet’s before you inject the air. Ideally closer to 50 is even better.
And then once you’ve created that bubble, you’re going to remove the stroma close, all the way down to the Descemet’s, and enter the bubble and remove the stroma as shown in F. And then basically remove the endothelium from the donor tissue. And usually we oversize the tissue by about a quarter of a millimeter. Some people recommend same sizing, you don’t want to oversize too much, otherwise you’ll get wrinkles in the Descemet’s membrane which can be visually significant for the patient.
Once you place the donor tissue on the eye, you can use multiple suture techniques. Interrupted, running, combined interrupted and running combinations, and all of them are all right. All of them are okay, it kind of depends on what the situation is, why the patient is having it. Because a lot of neovascularization, maybe the patient had an ulcer, you’re going to want to use interrupteds in suture. And this is an overview of all of them.
So the challenge is really for DALK is the big bubble formation without perforation. And you have to understand that perforation can occur at any step in the procedure. During the trephination, which we talked about, during placement of the dissector, I perforated using a dissector before I even placed the cannula needle. Again, if you have access to some of the newer technology like intraoperative OCT, that can be helpful. And then you have to determine what type of bubble you get. There’s different types of bubbles. A Type 1, a Type 2, a Type 3 and that’s important to identify as well. We’ll touch on that briefly. And then during corneal stromal removal, it’s very easy to perforate then, while you’re making the brave slash, or you’re removing the corneal stroma and we’ll talk a little bit more about that too.
So this is just a video showing what the ideal video looks like. You’re injecting air, you’re separating stroma from Descemet’s, which includes Dua’s layer. You can see the dissector and now the cannula. And as they inject the air, we’re going to create this nice bubble. The bubble is usually around seven to eight millimeters in diameter. You can see to the right limbal area, around 10 o’clock, there’s the bubble that’s been placed. You can see it’s displaced to the periphery, so you know the bubble centrally is displacing that air bubble peripherally.
And this is that bubble confirmation. You’re pushing the bubble to the periphery and here you can see that done very nicely in this video. You can see the bubble’s pushed to the periphery and then that’s the best thing that can happen. Then you can show, you can feel good and do a little victory lap there, with the air bubble going around the eye to confirm that the air bubble is there centrally, so that’s a very good feeling. That usually is why we do it, that’s one of the major steps to successful DALK.
The other issue is the brave slash. You have this air bubble, you’re trying to enter that bubble. And what we do is we place viscoelastic over the opening of the area we’re going to enter, and once we enter we’re going to get, the air’s going to come out. You’ll see the peripheral air bubble kind of moving centrally. Meaning that central bubble’s collapsed. Let’s just take a look at that again.
Let’s look at the brave slash again. We’re using a 15 degree blade to enter that bubble, that was one of the questions, one of the informant’s asked. And placing that little bit of viscoelastic, the HPMC or whatever it is, will work well. And when you enter the bubble, you can use a viscoelastic. Now this is important. Ideally, we want a cohesive viscoelastic that will easily be removed. Dispersive viscoelastics tend to stick around and we don’t want any viscoelastic left in the interface. So if you have a choice, we prefer the cohesive viscoelastics over the other. This is just a synopsis, quickly, very brief animated video.
So we’re making this trephination at two-thirds depth, we’ve marked the cornea, you can see the radial marks there. Because it’s so hazy, we’re actually and this cornea happens to be very thick, we’re taking the anterior lamellar off first before we do the dissection. There’s the 27 gauge Fogla dissector, there’s the 27 gauge cannula, and here we’re getting a bubble and we’re cutting into it a little bit. We’re putting visco, this cohesive viscoelastic in that bubble. Removing that stromal tissue, again this is just a standard Westcott scissors. You can see Bowman’s here, we’re removing Descemet’s and the endothelium on our donor tissue here, this has been quarter millimeter oversized, and then we’re doing suturing the lamellar transplant in place. You have to remember this is a combined interrupted and running suture combination. And you just have to emember you can perforate at any time. Usually if you perforate while suturing, that’s an easy, you can continue to do the surgery without worrying about problems.
Let’s talk about the big bubble. So this is a true or false question. In DALK, Type 2 and Type 3 bubbles combined are more common than Type 1 bubbles. And this is an easy question, true or false, in DALK, Type 2 and Type 3 together or combined, are more common than Type 1. It’s important to know what kind of bubble you get. So we got a 50/50 split here. But actually, Type 1 is more common. So this is going to be false. Basically Type 1s are more common, they represent about 80% of all the bubbles we get, fortunately! Because that’s the one that’s easiest to work with.
Type 1 bubbles basically represent, those are the classic ones that we see, they’re about seven to eight millimeters wide with a classic bubble. They start centrally and they progress peripherally. Whereas Type 2, so that includes Descemet’s plus Dua’s layer, which is about six or seven microns thick. Type 2 bubbles, actually, represent the splitting between Dua’s layer and Descemet’s membrane and you can see that here. It has a much more glassy appearance, you can see here in the lower slide to the right. And the bubble looks like it’s in the AC, but it’s not mobile, it doesn’t move around. And it’s much more common with scarred corneas. You can see the H and E staining here above the Type 1 bubble, you can see there’s an extra little seven, or six millimeters of stroma, that’s the Dua’s lager and then on the lower image here you can see just Descemet’s. So this is, Type 2’s actually very scary because the bubbles perforate very easily. And then you can have a Type 3, which is a mixture of Type 1 and 2, which is the least common.
Now there’s a new technique that’s been described for identifying what kind of bubble you have. It may sound silly, but sometimes you can’t tell how big your bubble is because it’s so white. You see this opaque bubble layer here above, where you can’t tell the size of your bubble. If you take a transilluminator of any kind and shine it over the bubble, you’ll be able to see how big it is. You’ll be able to see if you have a Type 1 or a Type 2, based on shadowing, and you’ll be able to tell if you have a Type 3, a combination. So this was just presented at the 2020 virtual ASCRS meeting, and I found it an extremely useful tip and I’m going to start doing this. I have not done this myself. I used transillumination for EMEK surgery, not for DALK, but I’m going to start doing this. This is a great tip that I just learned.
We’re always learning with DALK, we’re trying to pick up as many tips as possible. Fortunately the postoperative care of DALK is very similar to penetrating keratoplasty. We’re going to monitor patients for inflammation, infection, rejection, astigmatism management, as well as suture-related problems. Patients are going to be on long-term steroids and antibiotics. Now, some people believe you can take people off steroids quicker or more rapidly with DALK. That might be true for some type of DALK conditions, but in general, I personally believe people need to stay on a low dose steroid long-term. It may not be everyday, it might be every other day, it might be a few days a week. But we don’t want a patient to develop rejection. The benefit is that maybe sutures can be removed a little bit earlier with DALK than PK, which is a big advantage.
This is a video I call, “Avoiding the Mistakes.” There’s many, many, many mistakes that can occur with DALK surgery and this is a case I did and I made lots of mistakes. I’m going to share them with you right now.
One of the issues is when we’re doing, this is a keratoconic patient, he has an advanced cone. And sometimes that vacuum suction will not stick, you can try multiple times. And if you don’t realize that your vacuum trephine won’t stick, you have to turn the trephine barrel further back, much further back than you would expect in order to get suction. And as a new person either using vacuum trephines or doing a DALK surgery on keratoconics, you need to realize that you need to back up that suction. So here I am, trying, trying, trying to get suction and I can’t. So finally I realize I need to, but I’m thinking there’s an alignment issue, I’m thinking that my assistant, who has a syringe, they’re pushing the plunger in, they’re not doing a good job. But it’s actually me. I’m the one that’s responsible for having a problem. And then I finally realize that I need to back up the barrel in order to get adequate suction for this patient.
So that’s what we’re trying to do here. Seems like a simple thing, but you know what? If you can’t start the procedure, you can’t get started, you’re not going to be able to do the DALK procedure. So I finally back it up and got the appropriate depth. So learning how to use that vacuum trephine appropriately, particularly for keratoconus is very important. Again, and with this particular trephine, we get about 50 to 75 microns per quarter turn. So I’m counting as I turn and I back it up a lot, which is why there’s so many rotations there. We like to check the depth, I got great depth, we drive the edge here. And when I get to the bottom of that trephination, and then advance the dissector accordingly.
Again, one of the mistakes is not getting deep enough into the tissue. The deeper you are with your dissector the more lamellar the corneal fibers are. And it’s easier to actually advance the dissector the deeper it is in the corneal stroma. Here we’re trying to hold the tissue, you can see there’s a lot of distortion. I’m probably too anterior, because it’s so hard to advance to dissector. You need to be deeper. And it’s very hard to be deep, so we’re going to try again. And this time we’re going to try to be a little bit deeper.
Trying to get way down as deep as possible. Oh wait, this is the cannula, I think, right? And then once you advance the cannula, you need to advance it beyond where you dissected it. You see, I squirted water on there, once you squirt water on there it’s much easier to tell the depth of the cannula. You can start to see some folds or wrinkles in Descemet’s. And again, once I advance and get far enough I can do the injection. And I press down and you can see I got a nice bubble here.
Now you can see that the top part’s a lot more white, the other thing I’m seeing, I’m pressing on the side of the eye, the pressure is very high now when I create the bubble. So I just used a 15 degree blade to release some fluid. And now I’m going to take off the anterior corneal stroma. And where the part is white, it’s very easy to remove that tissue. Where there’s no emphysema of the air, it’s going to be a lot harder to remove that tissue. Now you can see, as I dissect down, you can see there’s a transition to a less white area. That means I’m getting very deep. So that tells me how deep I am. So I’m using that as a gauge and then I’m going to remove the rest of the corneal stroma.
Usually in keratoconics, I usually try to save the center for the last. Here I’m not sure if I did or didn’t. But it’s very easy to rupture, usually at the thinnest point, which is usually at the apex of the cone. There’s usually scarring there and it’s usually very thin. Most of these patients already have corneas that could be less than a 100 microns. But we take that off. Here I’m actually using, I’m injecting viscoelastic with a needle, as opposed to doing a brave slash. So this is the other technique that’s used. Problem with that is that visualization is hard and then usually when you cut down there’s already viscoelastic in here. This can be opened up then pretty easily and safely. You can see the viscoelastic coming out, the air bubble is released,and then we can remove the stroma. So that’s just two ways to deal with that portion of the surgery.
And then we put more viscoelastic in and then we go ahead and remove that. Starting removing the stroma’s not so easy. Most people do a cruciate configuration, meaning they make four slashes. Again, I’m always very nervous when I do this part. I don’t know quite how big my bubble is, so I’m going to inject lots of viscoelastic. Sometimes I go in with a dissector, just to make sure that the tissue has been released. You can sometimes have adhesions and if you don’t identify those ahead of time, those adhesions will keep you from… Will be a reason why you might rupture the bubble.
So again, I spend a lot of time taking this off. Some people are much more aggressive about it. But again, I’ve ruptured at any time. This is my second instrument that I was telling you about that I usually go around the edge. Just verify that there’s no adhesions, that there’s no, that’s it’s fully separated out to the trephination. Because if it’s not and I cut, I’m going to cut right through the Descemet’s membrane. And once I’ve done that, then I can take the rest of this off.
Once this part’s done, I usually put a little sponge over this area to keep it wet and I then prepare the donor tissue. We usually like to use trypan blue to stain the endothelium to remove it, so we can make sure that we remove it from the donor tissue. There’s been reports of people leaving the endothelium in place. Some have reported no problems, but obviously, most of the time we like to do that, remove. And again, trypan makes it very easy to remove the endothelium.
Once we do that, we have to make sure that we remove all the viscoelastic. This part is very important. If you don’t remove enough of the viscoelastic it’s going to get stuck between your donor tissue and the patient’s Descemet’s. And I usually dry, like I am here, and then I’ve already prepared the tissue and taken out the endothelium. And then I put that in place.
Now sometimes it doesn’t look like it’s the right shape and that’s partially because of the air. But once you start suturing in place, it’s very good. The other thing it’s important to know if you haven’t done a lot of DALK surgery, is that once you remove the endothelium, the tissue swells up dramatically. And when you’re suturing into a keratoconic cornea that’s thin, there can be a very big disparity. So the most important thing here is to remember you have to suture the tissue tightly. Much more than you would in a regular penetrating keratoplasty. Just because you have this very large disparity between the donor tissue thickness and the host corneal tissue. I will move forward here, I don’t think anything else here to talk about.
So let’s talk about some of the complications, but I’m going to ask one more question. One average, Descemet’s membrane rupture during DALK occurs about how much? 5% of the time, 15% of the time, 30% of the time, or 45% of the time? So on average, excellent, wow, great! You guys already know this. (laughs) So when I first started doing DALK in 2007, I think my Descemet’s membrane rupture rate was much higher. Definitely surgeon experience plays a big role. But it’s about 15% of the time that we can expect that. It doesn’t matter how experienced you are as a DALK surgeon, there’s always a possibility of rupturing Descemet’s, but on average it’s about 15% of the time.
There are specific complications we need to be aware of with DALK, intraoperative Descemet’s membrane perforation, the big one. You can have, if you rupture, you’re going to have collapse of the anterior chamber and it makes removal of the stroma tissue quite difficult. You can have double anterior chambers, usually because there’s a small rupture of the posterior Descemet’s membrane and there’s either/or retained viscoelastic in the interface as well. And you can even have air trapped between Descemet’s membrane and the host’s cornea.
Postoperative we have other specific issues such as stromal rejection, graft failure, infectious keratitis, Urrets Zavalia, and interface opacities, you need to know. So there’s been some reports out talking about outcomes with DALK surgery. This paper from 2007 in cornea, covered a bunch of issues. But if you look at complication rates, you’ll see that corneal perforation in this study was about 14%. Conversion to PKP was about 9%, a graft rejection around 10%, double AC is 4%, broken sutures are very similar to PK, 20%. Suture abscess 4% and steroid response about 4%. So again, perforation rates vary, anywhere from 9% to 40% based on the studies that you looked at. Some of the newer studies have lower rates, although the one from David Smadja is 27%. But intraoperative perforation, this is the one that people are most usually concerned about. And they occur more commonly in keratoconus, which is the most common indication for DALK.
Microperforation is usually that highest when you do manual dissection, it’s been reported around 26% and lowest with big bubble technique. Microperforation obviously can occur anywhere. Usually the times I see it, if it happens centrally early on, it’s very hard to manage. If it happens later in the procedure, let’s say while you’re suturing, it’s very easy to manage. So the timing of the perforation is very important.
Ideally, the other time that I sometimes see this happening is when people are removing the anterior stroma after they’ve opened the big bubble. And again, usually those can be managed, the peripheral microperforations very easily. Central perforations are harder, they tend to also rip or extend, which can be a problem. And depending upon when it happens and how large it happens, it can be an issue. Now there are many ways to patch these, either with stroma, sometimes people use the Descemet’s membrane with endothelium from the donor tissue. People use fibrin glue, a number of different techniques that can be used for this. Again, if you don’t manage these you’re going to have to convert to penetrating keratoplasty.
Pseudoanterior chambers, again, are not uncommon and those are usually due to breaks in Descemet’s or retained viscoelastic. And I’ve had to release sutures to release viscoelastic in some cases. Or put in multiple air bubbles to resolve the issue.
This is just a video showing sometimes how we have to break the bubble if we are too aggressive with our DALK surgery. You can see that here. Either the cannula was too close or the Descemet’s was too thin. Again, with the brave slash, we’re not using viscoelastic or if we failed to remove air from the anterior chamber beforehand, the Descemet’s will rise quickly and it will rupture. So in this kind of case, if you’re a beginning DALK surgeon, you’re done. You have to convert to PKP.
The other question that came up for people is what kind of anesthesia do you use with DALK surgery? In general, I know a lot of places favor general anesthesia for penetrating keratoplasty for a variety of reasons, I think that’s the best scenario to be learning DALK, is that you have someone under general anesthesia. But to be honest with you, here in the U.S. we mostly do this under local anesthesia. So just with a peribulbar block or maybe a little bit of a retrobulbar block for this procedure since it is considered a relatively extraocular procedure.
Rescue techniques, we talked about a little bit already. You can dissect around these perforation sites, you can use air, that’s probably the most common way for small microperforations. You can use a suture to close the tear, I’ve not done that, but it’s been described. And usually patch grafts or fibrin glue have been described as well.
Obviously the best medicine for this is to prevent rupture from happening. This means potentially using adjustable depth trephine, employing intraoperative pachymetry, using a pachy-bubble technique described by Ramon. I make some, but more common things is to make your paracentesis ports very peripheral. Also creating a tunnel for air, peripherally, if the cornea is very thin. And not giving, people tend to inject the air very, very forcefully, and that can create a lot of problems. So we recommend a relatively gentle injection of air. And you can in that way control the size of the bubble. You want to make sure you lower the pressure after you create the big bubble. And make sure that you dissect beyond the donor bed diameter to prevent rupturing.
Other potential problems include interface wrinkles, which usually have to do with steep corneas. You can have wrinkling peripherally but you really don’t want to have it centrally as it’ll affect vision. A graft rejection, in general, is about half of what it is for penetrating keratoplasty. Penetrating keratoplasty’s around 20-25%, where graft rejection in DALK is around 12-15%. Interface vascularization and opacification can occur as well as infection keratitis. There is a little bit more issues with fungal keratitis from the donor tissue, which is being addressed with some antifungals in the storage media.
You can see here, the picture is showing the inferior part of the cornea looking clear, the superior cornea looking hazy. This suggests that there is fluid or retained viscoelastic in the interface. Slit lamp magnification, you see there’s a little gap up here, indicating that separation. That could maybe be managed with air bubble, or if it’s viscoelastic, you might have to irrigate it out a little bit. But many times we can also just observe it, and if it gets better. In this case I think we have about 50% clearance, so that just might be an observation.
Vascularization is really a big problem with this and with lamellar surgery, so patients have a lot of vascularization, this is something you have to be concerned about in performing a DALK procedure. A little bit is okay, but if you have a lot of vascularization, it will quickly opacify the cornea and reduce the patient’s vision and survival of the graft. So we have to be careful in these cases. You can see this is probably, I’ve had a few patients have these problems and it does require, if patients stay on their steroids long-term they should avoid these problems. Buf if they become non-compliant or stop using their medicine it will be a bigger issue.
This is rejection. Rejection does not look the same as in penetrating keratoplasty. You’re not going to see KP, you’re not going to see a Khodadoust line, you’re going to see corneal edema and haze. And you have to just be aware of that. Increase being pentametry is going to be your best measurement with decreasing vision.
This is the last question. Which statement is true regarding DALK and PKP? Number one, the rejection rate and infection rate is the same for both DALK and PKP. The rate of rejection for DALK is less for DALK, but infection the same for PKP. C, rate of rejection is same for DALK but infection rate is less. And D, rate of rejection and infection is less for DALK. Great. Rate of rejection is less for DALK but infection rate is the same. Excellent, that is the correct answer.
Infection, unfortunately, is still about the same for DALK because we have sutures and anterior surface we have to deal with. But rejection is definitely less.
Stromal rejection, as I mentioned, it’s about half of that that we see with penetrative keratoplasty because we don’t have to deal with endothelial rejection, which is the most common type of rejection we see with PKP. And you can get, as I mentioned before, light sensitivity, redness, and corneal edema, and some vascularization. You’re not going to see that typical Khodadoust line or keratic precipitate.
Graft failure is usually less than with a PKP. Several studies to look at that. The Jones study’s a little bit of an outlier. But most people feel it’s definitely less. Sutures, unfortunately, complications are still common with DALK. I had one patient who got a suture abscess that actually went on to develop endophthalmitis from one of my DALK patients. So people can develop endophthalmitis with DALK as well.
Folds, I mentioned before. You want to keep them peripheral because they tend to occur because the corneas are very steep. And you can essentially massage them to the periphery. And you just want to get them out of the visual axis. Sometimes people readjust their cardinal sutures during surgery to minimize that. Urrets Zavalia usually occurs because we put air bubbles in these eyes. And the high pressure basically creates an ischemic event to the iris which causes it to dilate permanently. I’ve had one patient get Urrets Zavalia. In general, I’m putting much less air than I would in an endothelial keratoplasty, like a DSEK or DMEK. I try to leave the air bubble at less than 50% in these patients. We don’t need that high pressure that we need for endothelial keratoplasty. And I’ll skip that for now.
Outcomes, again, we kind of addressed. But you have reduced rejection rate, you have higher endothelial cell counts, you tend to have less astigmatism, in general, as you do with PKP. And fortunately, the visual results are very similar between the two surgeries. Probably the biggest issue is tissue dissection and surgeon experience in obtaining the big bubble. But unfortunately, the refractive issues are fairly similar where patients can have wide refractive errors after their procedures.
Survival, again, most people feel DALK out performs PKP with several studies showing good retention of the DALK at five years. And projected survival rates of up to almost 50 years with DALK.
Transitioning to DALK. probably the best thing to do is to practice your big bubble on all PKP patients who are under general anesthesia. Take that little extra time to get that experience from the real patient. Try to get a colleague to help with the first few cases. It’s best to do this as a team effort. You may have another colleague who’s trying to learn this technique. Obviously try to attend wet labs and courses at meetings to get all the tips you can. I’m still learning how to do this procedure better all the time. And review online videos and do not, do not give up. I think this is the biggest piece of advice I can give you. You’re going to fail. It’s not an easy procedure. But if you work at it, you will be able to master this procedure.
So these are the tips that I’ve learned. It’s easier to get a big bubble in an older patient. So if you have an older patient, let’s say someone who had a corneal ulcer that needs a big bubble, who’s in their 50s or 60s, it’s going to be easier to get a big bubble. Use a cannula, it’s much more forgiving and easier to get a big bubble than using a needle. You want to inject the air not forcibly. You want to do it gently or firmly, and try to control the big bubble formation. If you have access to a pachymeter in the operating room you can do pachy-bubble, that allows you to get deeper with your trephine. And in this study, in general, most people might get around a 65 to 75% bubble rate. Here they’re getting almost 90% if you’re able to do a pachy-bubble technique.
And then tie your sutures tight. I think a lot of people tie them too loose, I remember the first DALK I did, I tied it like I would a PKP. And at one week, I had to go take the patient back and re-suture the DALK completely because it was too tight. So I tied it much tighter and you want to obtain that 90% depth that you typically do with a PKP.
I always place a small bubble at the end of the DALK surgery just to help with Descemet’s. So those are the PEARLS that I’ve learned that I think are most important. And the other thing is, is the long-term, if the patient haven’t re-epithelialized by a week, put them in a bandage contact lens. There’s an issue, especially with keratoconics, with graft-host asymmetry, and it makes it hard for the epithelium to cover the new graft in the beginning. So use a contact lens or patch the patient.
This is a patient with mucopolysaccharidosis, preop and postop. And you can see the big difference in the quality of vision. This patient’s out now, probably almost 10 years, doing very well.
In conclusion, DALK’s experiencing renewed interest because it has advantages of both lamellar and full-thickness transplantation. We think, in general, DALK should be attempted as a first-line treatment for anterior corneal opacification and structural defects with healthy endothelium. We’ve seen there’s a lot of benefits to DALK with decreased rejection, it’s a closed-sky approach, faster visual recovery, quicker suture removal, and rapid steroid taper. But it is more time-consuming and technically more difficult. Although newer techniques have improved the predictability of this procedure.
The question is, how do you normally manage this vascularization post-DALK? So that’s a great question. Again, the best way is prevention. Making sure the patients understand that they have to stay on their steroids. So you can do, you have a few options. One is you can do a subconj kenalog or triamcinolone injection near the area of vascularization. If you have access to an anti-VEGF agent, such as Avastin. You can do an Avastin injection. And fine needle diathermy I have not done. The problem is, is that these corneas rapidly opacify in the face of vascularization. So even if you get rid of the vascularization, or reduce it, the corneas tend to remain opacified. So I’d have to say this is probably one of the most challenging situations. But I think the point of this is if you start to see this, be aggressive. Be very aggressive. Because once it opacifies, it doesn’t clear up. So it’s not like graft rejection that potentially is reversible, you might get the vessels to go away but the opacification is going to stay. And you’re going to essentially lose the graft, even if they don’t have a frank rejection episode, just due to the opacification.
So I think all of the things mentioned here as options are good options. And you can certainly combine them to basically encourage regression of the blood vessels. So again, thank you very much and it’s been a pleasure talking with you this afternoon.