Lecture: Glaucoma Filtration Surgery

In this lecture, Dr. Kahook covers the basics of filtration surgery with concentration on trabeculectomy. A discussion of Baerveldt implantation and the surgical management of tube erosion is also covered using video discussion. Audience questions centered on various management topics for blebs and complications.

Lecture Location: on-board the Orbis Flying Eye Hospital in Bridgetown, Barbados
Lecturer: Dr. Malik Y. Kahook, Professor of Ophthalmology, University of Colorado Anschutz Medical Campus


(To translate please select your language to the right of this page)

DR KAHOOK: These are my financial disclosures. And I do want to acknowledge several of my colleagues, because a lot of these slides were cobbled together from me and several of my partners. Leo Seibold put together a lot of these slides, and I also have a couple of videos from Sanjay Asrani, who’s at Duke, and Gayle Howard gave a talk at AGS on repairing the conjunctiva posterosions with a glaucoma drainage device, and we have a couple of those slides included in here as well. My first few slides are just for historical purposes. You don’t have to remember any of this. I just want you to see what we used to do from a glaucoma surgery standpoint, starting way back in 1858, with the anterior sclerotomy that was done. And this was the technique. You see it in the picture. Basically introducing a knife and then raising it to create a fairly big incision to allow for drainage of fluid. Now, eventually over time, like all surgeries, things evolved. Adding a miotic, daily ocular massage, added iridodialysis, and then this was largely abandoned by the 1900s, because of complications. You can imagine, by doing this full thickness procedure, that you would have a significant amount of complications. The irido-sclerectomy then followed by now not just working on incision within the cornea and the sclera, but also acting on the iris itself, to make an opening for increased drainage of fluid. And then eventually working itself to a full thickness trephination, a full thickness hole, which was one of the standards for glaucoma surgery for a very long time, in order to enhance drainage. As you can imagine, just like the first slide that I showed, a significant amount of hypotony, choroidals, complications that happen with that, because you don’t have the guarded filtration that we think of today, with a traditional trabeculectomy. Now, eventually other surgeons started to add their flavor to these procedures, and one example of this is the iridencleisis by Dr. Sugar. This was a procedure that worked fairly well. If you look at the data for how this worked, and basically it’s making a full thickness hole, pulling iris into that pathway, and allowing it to keep that pathway open, and in some cases, there’s some conjecture — not a lot of science — that it actually prevented the fibrosis from happening by having the iris in that place. This resulted in much better outcomes, but there was this fear of sympathetic ophthalmia, with externalizing the tissue, and this was also abandoned over time. Thermal sclerostomy, making that full thickness hole, and causing some cautery damage to the sclera in the area where the drainage was occurring was tried for a short period of time. Popularized by Dr. Scheie at the University of Pennsylvania. And all of these were associated with the common complications that you would think of. If I just asked you simply — if you put a full thickness hole into the eye, what are the complications you would expect, this is the list. Flat AC, iris prolapse, cataract formation, of course, with the shallowing of the anterior chamber, and hypotony, of course, because of the full thickness procedure that’s not protected. You can always get wound leaks. And then scarring and failure — you really went one way or the other, with many of these procedures. You had a lot of hypotony or you had scarring with failure and an increase in the intraocular pressure. And now we’re working our way towards the more modern trabeculectomy. This is guarded filtration surgery. Watson is the example here we’re showing you with the illustration. And the concept here was very similar to the full thickness procedures that I was describing earlier, but with a flap of sclera that can then be sutured in place to titrate the amount of fluid that comes out of the anterior chamber. This was not the initial intent of doing the trabeculectomy. The initial intent was to create the flap, go in, remove a portion of Schlemm’s canal and trabecular meshwork, and allow for flow to get into the cut ends of Schlemm’s canal, so that fluid would drain in the remaining Schlemm’s canal, and then go out the normal collector channel distal outflow system. Now, of course, we know now, from histologic studies, when we do a trabeculectomy and we do the tissue punch, we actually rarely get trabecular meshwork. We rarely open up Schlemm’s canal, and it turns out that this procedure was eventually modified so that the sutures in the scleral flap were the titrating factor, to allow for fluid to drain into the sub-Tenon’s space, and eventually be picked up by the bloodstream, and take the fluid away. And of course, there is this transconjunctival flow that happens in many patients, when they achieve a thin and cystic bleb. So trabeculectomy. This is what we do today, from a guarded filtration procedure. All of those that I gave for historical reasons are essentially not used anymore. The mechanism for IOP lowering is drainage through the fistula at the limbus, underneath the scleral flap, through the areas that are not as tightly closed with the sutures. Aqueous outflow is to the subconjunctival bleb, and it’s absorbed by the surrounding tissues. So the pocket of fluid is created, and eventually the surrounding tissue will then take away the fluid, and supplement for the diseased and blocked outflow system of the trabecular meshwork. The indications — typically we reserve this for patients who have moderate to severe disease, who have failed medication, failed laser, and of course, you heard me talking during the last surgery that we did, as well as in a previous talk during this week, that we’re starting to change the algorithm, that after laser, we’re going to the more microinvasive — microincisional glaucoma surgeries. But the traditional algorithm for trab and tube, waiting for the more severe disease, hasn’t changed much. We’re still reserving it for the more diseased eyes. We have a lot of clinical evidence, I think, to support this. This is changing over time, with a lot of studies like the trab versus tube trial, that gave us a lot of information on success rate, complication rates. The efficacy is very good. And my partner wrote this slide, and he says at least initially. Because this is what we say for all of our surgeries with glaucoma. Almost all glaucoma surgeries work for at least a little bit of time. Sometimes it’s five minutes, while the patient’s on the table. Sometimes it’s 15 years. But our surgeries do tend to fail over time. The safety is good without the antifibrotics, but since the introduction of antifibrotics like 5FU and mitomycin, while our success is fairly good, the complication rate has also increased. I do want to point out that the flow isn’t just through the scleral flap that’s been created and through the margins where the sutures are. You also have some amount of flow that might go into Schlemm’s canal, some amount of flow that might go through the sclera, and it’s not just that distal flow through the flap. So just keep that in mind. There might be multiple mechanisms for where the fluid is flowing, post-trabeculectomy. Now, this is the technique. I’m not gonna go through all of these steps. I just want you to realize that there is a cookbook recipe to how to do a trabeculectomy, and everybody starts off with the way that they were trained by their mentor, and then you can start to make it your own. So when I show a video here, I don’t want you to think that this is the only way to do it. There are multiple ways to do it. But you’re gonna see our technique. And for the sake of time, I might fast-forward through some of these. We always start off with a bridle suture, a 7-0 vicryl suture. We keep it double armed. We do a half thickness throw through the superior cornea, right against the limbus. What you’re not seeing right now on screen is we take a hemostat, we wrap it around the double armed suture, and then we fasten it to the inferior speculum, so that we can pull the eye down, and then with non-toothed forceps, you grasp the conjunctiva and start off with Vannas scissors, and you make a cut that’s about 3 to 4 millimeters along the limbus, in order to achieve an opening that would allow you to dissect. You can then take your Westcott scissors, go underneath Tenon’s, and continue to dissect not just centrally, but to the left and right of the opening, in order to get a pocket that is as diffuse as possible. Now, one thing that we should mention from a mitomycin standpoint: You have two different — and mitomycin, I should say, is the primary method that we use for antifibrotic. I can’t remember the last time that I used 5FU, honestly. But you can take the mitomycin and soak it within pledgets, and then put it over the sclera for a period of time. 2 minutes, 4 minutes. Or you could inject the mitomycin into Tenon’s capsule, superiorly, away from the limbus, at the beginning of the surgery, and it tends to diffuse over time, while you’re doing the rest of the procedure. We tend to inject our mitomycin now. It saves time. It provides the mitomycin in the area that we’re targeting. But know that you could do either/or. We then create the scleral flap, half thickness through the sclera, with a straight blade, and it always helps to have an assistant in this case. And the Weck-Cel allows for smooth movement of the blade. One thing I do want to mention here, for any of the trainees that are in the room, that it’s very important to paint, feel like you’re painting with the straight blade. You don’t want to do a saw technique, where you’re approaching the scleral flap and moving fast back and forth, without doing it deliberately. It should be a painting motion back and forth, to allow for that smooth bed and that uniform thickness of the scleral flap. So you work your way towards the limbus. Let’s see if I can fast-forward this a little bit. I’m not finding my arrow. I guess we’ll let it run here a little bit. So at the end here, we’re doing the flap. We put a little bit of viscoelastic in the anterior chamber, through the paracentesis. Now, some people choose not to do this at all. Just inject BSS. Some people like an anterior chamber maintainer in this case. For us, viscoelastic works fairly well. You go in with the MVR blade, give yourself enough of a pocket to introduce the punch — in this case a Kelly’s punch — will go in. You want to make two punches to give you enough space. Remember, we talked in one of the earlier slideshows that you need about 45 microns in order to be non-flow limiting. So once you get above the 45 microns, in this case, you want to get a little bit wider than that. This is about a millimeter with two punches. Iris presents itself. You do the iridectomy. And the pupil then reforms, once the iris falls in, and you suture scleral flap in place with 10-0 nylon. I like to suture the flap watertight, and then have control with laser suture lysis afterwards. I know with low access to lasers in some areas around the world, you can do a slipknot, and some of you are familiar with that. But the major message here is that you get a lot more control if you can close off the flow as much. I don’t want a lot of flow. Maybe a little bit of a trickle is okay. But if there’s too much flow, then I’m not gonna have a lot of control afterwards. And what I like to do is put enough sutures in that afterwards I can laser one, two, or three, after the first week postoperatively, once I’ve given the eye some chance to heal, and then control that flow as much as possible on my own. The conjunctiva is then closed with 8-0 vicryl, and what we like to do with the 8-0 vicryl is provide for two interrupted sutures, and if there’s a dogear, which is that area of folding of the conjunctiva, once you’ve done your first interrupted suture, you can then attach it to the initial rabbit ear, if you will, of the suture that you’ve left behind from the interrupted. So let me explain that a little bit more clearly. So once you’ve done the interrupted suture, and you’ve cinched down the conjunctiva with a 3-1-1 approach, you have this dogear that you can see here. And that you can close by not cutting the suture after you’re interrupted, but taking the needle and running it two more times through that open area, and then attaching it to the proximal end of the suture that you started off with, and that tends to bring the conjunctiva down in a fold. You’ll see how it kind of comes down and folds over itself, and that’s much more comfortable for the patient, because now the knot is covered by that fold of conjunctiva, and you do the same thing on the other end. Frequently, on the other end, you don’t have to close the dogear, because you have it much tighter because of the first interrupted suture. One thing you can do, if you have an assistant, is have them grab the conjunctiva, and as you’re cinching down your 3-1-1, pull the conjunctiva down towards the center, the visual axis, and that gives you a much tighter closure of your conjunctiva. In this case, we didn’t do that but you can see that we still got the straight line of conjunctiva. And that’s one thing I look for in all of my closures, whether it’s a tube or a trabeculectomy. I want my two interrupteds on either side, and I want that straight line of conjunctiva going across. I don’t want it to mirror the limbus. I don’t want it to curve around the limbus. I want that straight overhanging line, and that tends to heal very nicely, and rarely leaks when you make this type of closure. Now, from a follow-up standpoint, I think a lot of us who are doing glaucoma surgery — you know, what I just showed you was the easy part. The surgery itself is the easy part. It’s really the management postoperatively, and knowing when to do something and when to watch… We typically do prednisolone acetate, Q 2 hours. Vigamox 4 times a day, and we do that for about a week. We ask the patient to wear a shield, not to rub the eye, to do most of the postoperative management that you’re familiar with. We like them not to exercise vigorously. And then if we need to, during the follow-up period, if the pressure is too high, or if we want the pressure to go lower, to a target pressure of 10 or 12, we start doing laser suture lysis, in order to titrate the pressure as much as possible. And you see the settings, 50 microns, 0.1 seconds, 300 to 600 milliwatts. And it’s nice to use a Hoskins lens, in order to depress the conjunctiva and have great visualization of that suture, before doing anything to it. Now, one thing that I have a much better appreciation of after my week here in Barbados is how different populations have different Tenon’s capsule thickness. The patients that I’ve operated on here uniformly have about a millimeter-thick Tenon’s capsule, which is not my normal — that I see when I do trabs and tubes. But in that case, you might have great difficulty seeing the sutures through the Tenon’s capsule, and one way to get beyond that is to make sure that the sutures that you’re putting into the scleral flap should be long. Don’t do the small, 1-millimeter course sutures, because they’re gonna be much harder to find. Give yourself a longer target, and when you laser, laser right in the middle of that suture, and then oftentimes what you’ll see is the suture will present itself by raising up. You should pay attention to that. If the cut end of the suture comes up straight towards you, and it’s looking right at you, that might cause some problems down the road, by eroding through the conjunctiva. Since you’re there anyway, take the same laser, and laser the distal — the cut ends of the suture — and have it shrink down and ball up. And that will never be a problem if you do it. So I routinely do the laser suture lysis. I cut it in half, and then I laser each end to make sure that it shrinks away from Tenon’s capsule, shrinks away from the conjunctiva, and it has a dull end to it, rather than a sharp tip that might come through with these 10-0 nylon sutures. So who is high risk for failure? In our case, in my patient population, the African-American community tends to have a higher failure rate. We tend to use higher doses of mitomycin C, and maybe have the exposure for a longer period of time. Instead of 2 to 3 minutes, it might be 4 to 5 minutes. Patients who are younger, of course, uveitic, active uveitic, active neovascular glaucoma patients, we tend to go more towards tube rather than trab, because of the high rate of failure. Prior failed surgery with conjunctival scarring and especially the prior conjunctival surgery — so sometimes you can get some fibrosis and changes to the conjunctiva, just from anterior chamber surgery. But certainly if you’ve had a trabeculectomy done before, or any type of injections done into the sub-Tenon’s space, those patients are more likely to fail, and might benefit from a higher concentration of mitomycin and exposure for a longer period of time. Anybody who’s been on drops for a very long time, which is why you heard me narrating during the surgery that I try and do everything possible to get patients off of drops… Most of the drops that we use can injure the conjunctiva. We know from multiple papers done long, long ago by people like Mark Sherwood, who studied histologically the tissue in patients who are on therapy or not on therapy. So patients who went for primary surgery, never having been on drops, versus those who were on chronic drops, and went for surgery — certainly a proliferation of inflammatory cells in those who were on glaucoma drops, and the preservative that we use in drops, benzalkonium chloride — in most of our drops, that’s what we’re using. And it’s much like a soap. And that can irritate the conjunctiva and cause inflammation over time. It also tends to accumulate in the Tenon’s capsule and conjunctiva. It doesn’t just wash away. You can go 14 days later, after a dosing, and still see remnants of benzalkonium chloride. So whatever you can do to get them off. You might choose, before surgery, in patients who are on two or three drops, to take them off of their drops and put them on a mild steroid, while they’re waiting to go into the operating room for their trabeculectomy. Try and quiet down the conjunctiva as much as possible, when that’s practical. Sometimes it’s not practical, because you’re going to surgery the next day. But if at all possible, honeymoon off the drops. Start a mild steroid. And try and get them as quiet as possible. 5FU is one of the antifibrotics that has been used historically. It affects DNA replication. There are multiple studies that were done, most of them small, but one of the larger studies, the FFSS, the Fluorouracil Filtering Surgery Study — 213 high risk trabs. To me, the data for this study… It wasn’t actually very compelling to start using it, once I was out in practice, which was much longer after the study was completed. At 5 years, failure was 74% in control, 51% in the 5FU eyes. Not very compelling for me to adopt this. Plus a lot of these cases with 5FU, it’s not just the intraoperative application, but a lot of postoperative injections that take place. For my practice, it’s not very practical. It really stops the clinic and can slow us down a great deal. Now, if it were very effective, of course we’d all make that sacrifice and do it, but it wasn’t very effective. Adverse effects including corneal toxicity are well known. It can also increase your rate of leaks. Mitomycin is probably globally the most common antifibrotic that people reach for. This is an alkylating agent. It crosslinks DNA. It’s non-specific. It also inhibits fibroblast growth and proliferation, and we talked a little bit earlier about how the application — you could be soaking it in a pledget, putting it on the sclera for 2, 4 minutes, and then washing away, after taking the pledgets away. The dose varies. We have eight people in our practice, eight glaucoma specialists. We all do things slightly differently, even though we try to be uniform. Not just the dosing, but also the duration. Multiple studies have shown that the risk of failure is decreased, but also the risk of complications are increased, as you might imagine, with an antifibrotic. Avascular blebs — they can leak late. You can get hypotony. You can get an increased risk of infection. And if you just look at the survey that the American Glaucoma Society did of what the usage pattern is, by far the large majority of surgeons are using mitomycin on the majority of their trabeculectomies. It’s not very selective as to who we use it on. But it is selective as to the duration of exposure. So if somebody has a very thin Tenon’s capsule, we might do an exposure for 2 minutes, instead of 4 minutes, and then the opposite if it’s a younger patient with very thick Tenon’s. More recently, there have been some FDA-approved, regulatory-approved versions, like Mitosol. This is a packet that comes out, where it’s ready to be mixed by the OR technician, OR nurse. The problem with this is it’s very expensive. I can get mitomycin made up in our pharmacy for just a few dollars. When this first came out, it was hundreds of dollars, and I think it’s gone down. I’m not sure what the final price has been, but it’s certainly more expensive than what I can get from our pharmacy. Wound modulations is necessary, but you just keep in the back of your mind… I’m using the mitomycin, so I’m probably gonna see more complications. But I’ll see less scarring, so to me, it’s worth that, where I fight to get the best amount of IOP lowering, but I do have to pay a cost, and the patient has to pay the cost of potentially having higher complications. Ideally, we would have scarring decrease and complications decrease, but we’re just not there yet. And I also think that we just… We haven’t gone that extra step, when it comes to wound modulation, where, if you look at other branches of medicine, we don’t use one drug for everything, like we do with mitomycin and scarring. And I think a cocktail, a mixture of drugs, to decrease fibrosis is something that we need to take advantage of from other branches of medicine. So if you look at cancer therapy, most of those patients are taking multiple medications. If we can somehow study the effects of mitomycin combined with other antifibrotics — we’ve done a lot of studies looking at Avastin and anti-VEGF agents, which also influence fibroblast growth, and trying to see if that has any synergy with mitomycin. The problem is it’s very hard to study. It takes a lot of money. It’s not… There’s no incentive for any of the big commercial companies to do these studies, and we’re left to try and learn from our own experience. I tend to mix mitomycin with triamcinolone, with a steroid injection, in patients who have an angry-looking conjunctiva. Prior to closing the conjunctiva, I take a cannula, and I go as far back as possible underneath Tenon’s capsule, and I leave a little bit of triamcinolone to calm things down after the surgery, and I do that in combination with mitomycin C, especially in eyes that have angry conjunctiva or younger patients. That tends to work fairly well. Now, I’m happy that I have that flow going through the trab, so I’m not as worried about a steroid response glaucoma, because I have control to laser some of the sutures, to open things up, if the pressure starts to creep up. There’s this common debate of limbus versus fornix. Just out of curiosity, how many people in the audience have done a trabeculectomy? So we’ve had a few. So raise your hands up who’ve done a trabeculectomy, all of you. And who has done — if you’ve done a limbus-based, keep your hands up. And if you’ve done a fornix-based, keep your hands up. So hands are going up and down. You’ve got both hands up. You’ve got a lot of experience, right? So it really depends on how you were trained. For limbus-based, better wound closure, less early leaks, avascular blebs. They leak late. Potential for more infection. Fornix-based — trickier wound closure. More leaks in the early postoperative period, if you haven’t closed your conjunctiva correctly. Better bleb morphology. More posterior. Maybe more hypotony. So it really just depends on how you were trained. I’m not gonna spend too much time talking about Ex-PRESS. I might show a video here, if we have time. How many of you have done an Ex-PRESS implant? Just out of curiosity. It’s not something that has made its way out of the US very much. So none of you, really. The Ex-PRESS device is expensive, relative to a standard trabeculectomy. In the US, the cost for us historically, just checking a couple years ago, is around $800 to $900. So adding that cost on top of some of the care that you’re doing, in areas where the patient might not be able to afford it, this is covered under insurance, and under our government insurance, so the patient isn’t paying out of pocket, but it’s still a significant burden on health care. So you have to weigh the pluses and the minuses. I will tell you that I do like using this in cases where I’m teaching, because it’s much more controlled. You’re not putting a hole in the eye with the punch, and struggling to do the sutures before viscoelastic is leaking out. In this case, you have a 50-micron hole in the device that is right on the border of being flow limiting, and it allows you a little bit more comfort. Once you put it in, you have some time to do the closure. So if you’re teaching a resident or a fellow who’s new to the surgery, it’s much more comfortable, and these patients also tend to recover their vision faster. So if you have somebody who’s monocular, somebody who is very dependent on getting back to work sooner rather than later, patients, post-Ex-PRESS, they do recover their vision faster than a standard trabeculectomy. Minor differences. Instead of doing a tissue punch with a trabeculectomy, as you see on the left, in the case of an Ex-PRESS, you do a 25 or 27-gauge needle sclerotomy. You implant the Ex-PRESS, and then everything else is the same. You suture the flap down. You suture the conjunctiva down just the same. Entry for this device — you do it right below the transition into the limbus, right posterior to the transition into the limbus. And let me show you a video of what that looks like. So the needle goes in. Really important when you’re doing this — same thing with the glaucoma drainage device — when the needle is coming out, try to stay in the tract. And in the case of an Ex-PRESS, we like that tunnel to look exactly like this, where there’s been no movement to the left or right, so that you don’t get the peridevice flow that can happen if you widen the wound. Now, with a glaucoma drainage device, if we get a chance to show that today, I actually do intentionally sweep to the right or left, so I can give myself some room to insert the tube. So I’m gonna skip to the next. This is what the insertion device looks like. It’s basically just depressing on this button, which bends this wire. When you bend that wire, the wire that’s going into the device basically shortens, and the device is left in place. This is what it looks like. You put the device on the side. Go into the tract that you created with a needle. Rotate it 90 degrees to get it even with the sclera. And then check for flow. And when I check for flow, I make sure that I’m checking not just through the middle hole. I’m checking around the device as well, to make sure that all of my flow is coming straight from the hole, as you see right here. And then after that, my closure is the same. 10-0 nylon to the flap. 8-0 vicryl to conjunctiva. Complications look very similar to what you would see for a trab. The main advantage, like I said, is that the vision recovery is much faster. But otherwise it behaves very much like a trabeculectomy. So a bunch of photos here that show you all of the reasons why we’re trying to transition to less invasive surgeries. Right? So this picture montage of different complications of blebitis, of leaks, of hypotony — we’re living in this zone, when it comes to glaucoma filtration surgery, and we’re trying very hard to avoid the currency of these. And, of course, the dreaded complication of endophthalmitis. Wound leaks are the main concern in the — not just the early postoperative period, but postoperatively, months, years down the road, when the blebs become thin and cystic. There are different techniques that you can use. If it’s just a wound leak, and you’re not worried about some of the issues that might rise from infection, you can use antibiotics as a precautionary measure. Pressure patches, large diameter bandage contact lens, which I’ve asked a lot of questions about during this week. It seems very hard to find in certain places around the world. But we have access to a bandage contact lens that we put on, when a patient is leaking. Especially when it’s fresh, after surgery. And that can tamponade things for some time, until healing can take place. You can use various glues, cyanoacrylate, not very expensive. Many places around the world have access to it. Fibrin glue tends not to be a good choice, because you need a dry surface in order to apply it. It doesn’t last very long. So cyanoacrylate would be the one choice, if I were to use a glue. And then you can always take the patient back into a procedure room, or back to the operating room, to suture and have it closed. This is from an old textbook, just showing different pressure patches that you can do in cases of hypotony and leak. Early postoperatively, it can happen when you have hypotony, it can happen from a wound leak, it can happen from excessive filtration, where you have a large, elevated, diffuse bleb. It can happen from ciliochoroidal detachments, where you’re getting hypotony because you’re not producing as much fluid, as much aqueous humor. It can also happen from a cyclodialysis cleft. And this can be just after what seemed like a routine phacoemulsification, where you can get a cyclodialysis cleft. It can certainly happen when you’re doing a trabeculectomy, that you induce a cleft, typically underneath the wound where you’re doing the punch, and the patients could get hypotony from that. And the clue for that, of course, is that you have a low bleb in this case, because the fluid is actually draining into the cyclodialysis cleft itself. In the late postoperative phase, hypotony is typically because of bleb leaks. Hypotony with deep AC — and I’ll just go over these slides, just different scenarios, fairly quickly here, and try to get to a couple of videos, before the time runs out. Hypotony maculopathy, fine macular striae radiate from the fovea. You see a picture. Choroidal folds, tortuous retinal vessels, and optic nerve head edema. This is the classic picture of hypotony that can be seen after surgery. Patients at high risk tend to be the younger patients, myopes, and those who are on a preoperative CAI. Typically my worry is young and nearsighted. Very nearsighted. Those are the patients that I worry about, and I really make sure that I’ve closed that flap very tightly. Vision loss is reversible, and the amazing thing — we’ve seen patients who have had hypotony maculopathy for years, and we elevate, we increase their intraocular pressure by modifying the outflow pathway of the aqueous humor, and these patients tend to get vision back, even years down the road. So don’t give up on these eyes, if they’re coming in, and they might be 20/70, 20/80, and their pressure is 4, and you can tell that the hypotony maculopathy, from exam, is what’s limiting their vision. Try and increase their pressure. There’s still hope for this eye to see much better. Hypotony with a flat AC, when you’re seeing that, look at the bleb. Is it flat or high and diffuse? Do a Seidel test to make sure that there isn’t a leak. And then determine the degree of flattening, and that will help you determine what your intervention should be. Shallow with no touch — and by no touch, I mean no iris or lens touch. You can just observe. You might start the patient on atropine, 1%, QID, deepen the chamber, and just watch for some time. If there’s shallow with iris touch, then you can also try conservative management, but you’re starting in the back of your mind to think: I might have to do something a little bit more, because that third stage of having lens touch is something that you have to act upon. One step that you can do in the clinic at the slit lamp is, if you have access to a paracentesis, and I do a paracentesis with all of my filtration surgeries just for this reason, you can inject some viscoelastic into the anterior chamber, deepen the anterior chamber, and see if you can buy some sometime, until the eye starts to heal from a bleb standpoint. And that’s extremely important, to try and buy some time in that way. I will say this: If you have an injection of cohesive viscoelastic, and you see the chamber has deepened after the slit lamp procedure, don’t just have that patient go home. Because these patients can spike significantly, and be in a lot of pain, with corneal edema. So you always keep these patients for at least an hour and recheck. If you’ve rechecked after an hour and they’ve shallowed again, you can try another injection and wait another hour. But if they’re holding steady and their pressure is not very high, you can have them go home and see you the next day. If you’ve rejected after the patient has shallowed once, post the first injection, I’m gonna tell you that the injections are not gonna help. So that patient has to go back to the operating room and have the filtration surgery revised. Overfiltration with flat AC. You can try atropine, decreased steroids, pressure patch. We went over a lot of these before. You can reinflate with viscoelastic. In some cases, you can use gas, which I don’t tend to do, and then the revision. There are other bleb-limiting techniques that you can do. And the most common one, I think, that’s done in most practices — certainly in my practice — is the compression sutures, where you can do this in a procedure room. Some people even do it at the slit lamp. And that’s putting long sutures, where you’re taking a bite through the conjunctiva, through the sclera. And then fastening it to the limbal area, so that you’re restricting the amount of flow to either side of those sutures. You’re basically remodeling the bleb with suture. You can also do a compression suture over the scleral flap, in case of a trabeculectomy, through the conjunctiva, through the sclera, and bridge the scleral flap and tie down on the opposite side, which will flatten the scleral flap, and hopefully increase your intraocular pressure by decreasing the flow. So there are some things that you can do. So for the sake of time, because I do want to get to a couple of the other procedures, I’m gonna jump forward and show you a little bit more about glaucoma drainage devices, with a video. And just show you the technique that we do for that. So this is our service at the University of Colorado. Leo Seibold. Jeff SooHoo. And our new partner, who’s our current fellow, Kara Capatina. So you’ll see a lot of similar maneuvers for trabeculectomy. In this case, we’re putting a bridle suture, 7-0 vicryl, and we’re fastening down with the hemostat around the speculum to infraduct the eye. We cut the conjunctiva, and we’re using the same type of caution that we use with a trabeculectomy. We lift up. We cut with the Westcott scissors. And in this case, instead of the 3 to 4-millimeter peritomy that we do with the trab, in this case, we’re doing a much wider conjunctival peritomy, and I also tend to put two 3, 4-millimeter relaxing incisions in each quadrant, to really get great exposure to the sclera. In this case, here’s that relaxing incision that I was talking about. And I do that on both sides. And then if you go in with curved Stevens scissors — those of you who have been in the OR with me, you’ve heard me say this several times — the curved Stevens scissors can be your best friend to go in as far back as possible, open up the scissors, and bring out the scissors while it’s still open. That’ll dissect much of Tenon’s capsule away from the sclera, and give you this wide opening, in order to be able to implant, in this case, what’s gonna be a Baerveldt implant. You hook the muscles, superior and lateral recti, tent up the conjunctiva and Tenon’s, and while you’re doing this, you’re also tenting up the muscles, whoever is assisting you is tenting up the muscles, and that allows you to slip in the Baerveldt with a little bit more ease. Now, there’s the argument of 250 versus 350. In my experience, 250 and 350 Baerveldts perform exactly the same. There’s a tendency to have more diplopia in our hands with the 350. So we’ve gone almost exclusively to just using the 250 Baerveldt. You then put sutures in, about 8 millimeters, if at all possible, from the limbus. There are some cases where it’s not possible to get back. 8 millimeters, if it’s a tight orbit, small orbital anatomy. And you tie it off. And now, the suture that you use varies by surgeon. You can use something that is non-absorbable, like a nylon or a prolene suture. I tend to use 7-0 vicryl for everything. My bridle suture, fastening down the plate, my figure of 8 suture over the tube, and then closure of my conjunctiva. I like the simplicity of it. I’ve never had any problems with moving or early biodegradation of the suture. And then you tie off the suture, because the Baerveldt is a non-valved implant. You can tie it off with 7-0 vicryl, and that’ll have a tendency to last about 4 to 6 weeks or so, before it opens up. If this is a patient where there’s higher pressure, and you’re worried about the pressure going too high, while you’re waiting for the suture to open up, you can take a 30-gauge needle and put it through and through the tube, far from where you’ve tied off the tube. Between the tube and the anterior chamber. That creates two openings on either side of the tube that will be a valve release for fluid, if the pressure goes too high. Put some viscoelastic in at this stage. Go in with a 23-gauge needle, 2 to 3 millimeters away from the limbus. If you can go 3 millimeters away, that’s gonna give you a lot of coverage over the tube, and it tends to lay flatter on the scleral surface. And in the case of a phakic patient, or where the anatomy doesn’t allow, we go into the anterior chamber, above the iris. If the patient is pseudophakic, we try to go into the sulcus. Depending on where the tube is, if it’s in the anterior chamber above the iris, we bevel up on the tube. If it’s going into the sulcus between the iris and a PCIOL, we do bevel down, and we have the bevel eye right on the IOL surface. And the tube hand over hand, with tie forceps, is the easier way to do that. So introduce it into the sclerostomy, and feed it in. You can see it sitting there in the anterior chamber. And then it’s followed by figure of eight suture over the tube, to tie it in place. You can do whatever type of suture is most comfortable for you. And then tie it into place again. I tend to use 7-0 vicryl for this. In some cases, some of my colleagues use a different type of glue, a Tisseel glue, which is a fibrin glue. And those patients also do very fine. So it’s just an indication that you don’t have to have something permanent there. And then Tutoplast goes over. We cut it to size, so that we’re covering the tube. We try not to cover the suture that ties off the tube. We try to leave that as exposed to the environment as much as possible so that it’ll loosen over time. And then you can either glue this in place — but that’s a little bit expensive. In this case, it’s the fibrin glue. You put a few drops of fibrin glue on the scleral surface. Tutoplast then goes over, and then you put glue on either side of the Tutoplast and over the Tutoplast. And then bring the conjunctiva down, and hold it. And you hold it in place for a little while, and it tends to hold very well. And this also in our experience decreases the amount of suturing that we do, and it increases the rate of healing. The patient tends to recover faster. Less injected, less inflamed. In some cases, where there’s a little bit of retraction, if there’s a little bit of scarring, we’ll still put in a couple of the interrupted sutures, in order to achieve full closure. So we’re approaching the time limit that they gave me for this talk. I know we were only able to cover the video for glaucoma drainage devices. I’m gonna take a few questions right now, whether it’s from the audience or from… Okay. Okay. I’ll do a little bit more, and then we’ll do some questions? Okay. Well, then, let’s talk about tubes some more. I’m sure we’re excited to hear that, right? So there are different tubes that we can use. And in the US, the typical tubes that we’re using include the Molteno, Baerveldt, and the Ahmed. In different areas around the world, you have access to other devices. Is it the AADI device? Is that the name? AADI device is from India. That’s very much like the Baerveldt device. Non-valved. But you can see some of the dimensions here are just good to keep in the back of your mind. You’re not gonna memorize all of these, but it just gives you an idea of what kind of space is available to you, once you’ve implanted these devices. Superotemporal, superonasal, inferonasal. It’s different depending on where you go. And I do want to leave you with that thought, when it comes to drainage devices. If you’re operating on a short eye, a hyperopic, short axial length eye, and especially if you’re going into the inferonasal quadrant, let’s say you put in a superotemporal tube already, or the superotemporal quadrant is not available, because of scarring, and you’re choosing to go inferonasal, you don’t have as much room from the limbus for the implantation process, because if you go on a short eye, if you go 10 millimeters back, whether it’s inferonasal or superonasal, there’s a risk of the distal end of the plate rubbing on the optic nerve. And that’s an important thing that I think is underrecognized. Sometimes when you put a plate in, and you put it way back on the sclera, and that distal end of the tube is rubbing against the optic nerve, it can increase your scotoma. It can look like a glaucomatous visual field defect that’s getting worse, and really what you’re doing is a compression retinopathy. So always look at the length of the eye, if at all possible, and keep in mind that if you’re going superotemporal versus superonasal, if the axial length is 22.5 versus 26, so just look at the superonasal example here. 22.5-millimeter axial length. 8.5 for the Baerveldt. 258 for the 350. That’s how far back you can go. But if you go to a 26-millimeter eye, you have 11 millimeters and 10.5 millimeters before the device is abutting the optic nerve. So it’s not really a one size fits all with these. You have to keep that in mind. Non-valved techniques. I talked a little bit about tying off the tube with a 7-0 vicryl. When I do that, if the pressure is in the high teens or low twenties, I don’t worry about doing slits in the tube as a release valve. I’ll just tie off the tube and then ride it out on the patient medications, until the suture opens up. In some cases, with very high pressure, you should take a 30-gauge needle or a straight blade and make slits in the tube, so that there’s a place for the fluid it drain into. You also have the option of doing what’s called a rip cord. So you can take a permanent suture, put it into the tube through the plate side of the tube, tie it off with either an absorbable suture or a non-absorbable suture, and then at a later time, you can either wait for that absorbable suture to go away, and you have perisuture flow in this case. If you’re putting the rip cord in the silicone tube. Or you can tie it off with a nylon suture, and come back a few weeks later, and laser that nylon suture, which would then open up flow. Because you have the rip cord in place, you’re not gonna get a gush of fluid. It’s gonna leak around the suture. So you’ll have more controlled decrease of your intraocular pressure. And you can always go in and take out that rip cord, and you can do this at the slit lamp, if you have a long enough length of your rip cord. Make a small incision in the conjunctiva. Grab it with jeweler forceps. And just pull it at the slit lamp, and it opens the tube to its 330-micron opening that it exists in. So there are different ways to control. I like the simpler route, when it comes to anything where I have a complicated on one side, simple on the other, I try and fix simple whenever possible, and that’s why I like to suture with 7-0 vicryl, tie it off, and I just put a release valve with a 30-gauge needle through the tube, for simplicity. Tissue reinforcement is a topic that varies widely in different regions around the world. In the US, we tend to use patch grafts. We tend to use Tutoplast that we purchase. This is processed pericardium. You can use processed sclera or processed cornea. And we put it over the portion of the tube between the limbus and the plate. This is expensive. And it’s not something that everybody has access to. So one other technique that you can use is: Go back 4 millimeters posterior to the limbus while bending your needle in a Z formation. Some of you might have seen this. This was something that was popularized by Felix Gil out of APEC in Mexico City, and he’s trained multiple fellows, who have gone on to train their own fellows in this technique. If you can get back 4 millimeters from the limbus before putting the tube in, the coverage tends to be adequate. You don’t necessarily need to put a patch graft on. It is, however, difficult to do for those who haven’t been trained on it, and it’s also more difficult to advance the tube through that long of a tract within the sclera. It’s much easier to do it if it’s only 2 to 3 millimeters away. But you do have options. One thing that came up in our discussion yesterday, I believe it was, from Dr. Sugrim, was: Can we just do a scleral flap and put the tube under that? And certainly you can do that. The problem there is that: Once you create the flap and put the 23-gauge needle through the tissue, in order to introduce the tube, that needle pass tends to be a little bit more irregular in allowing for peritubular flow. So it’s not as tight, and you can have a tendency to get hypotony afterwards. Complications post-tube: They look very similar to what we were just talking about, about trabeculectomy. You can have all of the same things, like hypotony. In this case, if the tube scars over, you can get what we call a hypertensive phase, which is exactly what you might get with a trabeculectomy, where the scleral flap just fibroses down, the pressure just goes up. You can get migration, erosion, extrusion — all of these things can happen. Corneal decompensation is slightly unique here, compared to the trab, because you have a device that could be rubbing on the cornea. You can get things like diplopia and infection. I’m gonna skip here a little bit to a video that shows tube erosion. So there’s an erosion here, right at the limbal area, where we’ve done a full exposure. And we have to fix this. So we go in and we make a peritomy. And we make a pocket over where the old Tutoplast was, but it’s no longer there. It’s basically melted over time, which we see, unfortunately, with the processed pericardium. We then take another piece of Tutoplast, and we fit it so that it can go into that pocket that we’ve created. We try and do minimal dissection, if at all possible. So we don’t lift up entirely and invade the bleb area that is now fibrosed around the plate. Once we get it into place, that area that had the previous erosion in it is cut out, and we get rid of the tissue in that area. So as much as possible, we try and preserve the conjunctiva, but we take out the area that had the hole. And we basically lift a hood of conjunctiva over the Tutoplast and resuture the healthy conjunctiva proximal to the limbus. In this case, you can see we’re using that Tisseel glue that we have access to. It makes it much faster. Less inflammation. And it holds both the Tutoplast and the conjunctiva in place, and you can augment it with sutures, which is what we see here. Very similar to what we described earlier with the 3-1-1. Now, here’s a slightly different technique. See if we can play this. This is Dr. Sanjay Asrani, doing his closure, posterosion. This is real speed. He’s just much faster than I am, as you can see here. And one thing that he does here is he cleans the epithelium in the area. He scrapes away the epithelium that might have grown down over the conjunctiva, into the hole that was present. And then his technique in this case is to put a piece of patch graft over the tube, and to close the conjunctiva, and not necessarily close it so that the ends approximate. In this case, if you close it and it’s close enough, eventually the epithelial cells will grow over the patch graft. And these patients tend to do very well. So it’s just another technique that you can use, if you’re getting into an area where you just don’t have enough conjunctiva. Now, I want to stop here and just mention a couple of things that I think are really important. Once you’ve lifted up the conjunctiva, and you take a look at the tube, in some circumstances, the tube just wasn’t put in the right place. It might be a millimeter from the limbus, which is too close. The plate might be too close to the limbus as well. In that case, I don’t do these techniques. I actually take the tube out. I suture the sclerostomy site, to make sure that I’m not getting any flow. Because now, that’s a very mature filtration tract, and if you don’t close it, the patient will continue to leak from that side. You close it off, and then you take a 23-gauge needle, and you make a path in a different area. You just move the tube to a different area, preferably to an area where you have a lot more conjunctival closure, and you do the closure in that area. That’s also another technique you can use if you don’t think you’re gonna have enough it conjunctiva to cover over that spot. You can just do a wider peritomy, direct the tube into another area, put the Tutoplast patch graft over the tube, and then completely close off the old site. There’s a lot of advantages to getting away from areas that had a hole and going to a very healthy area of conjunctiva. So diplopia is something that can certainly happen. I’m thankful that I don’t see a ton of it, and I don’t know if it’s because I use more 250 Baerveldts than 350, but there is some signal between my earlier practice and my current practice of using 250 versus 350. But it’s certainly something that can happen, and it’s something you should keep your eyes on. These patients can commonly benefit from seeing strabismus surgeons and getting prisms in their glasses. We are learning a lot about how a tube performs, versus how a trab performs. The TVT trial, and we’re getting a wealth of data, we’re getting five-year data right now, that you see in front of you — that tells us how we should expect, what we should expect from a trab versus a tube. And it could be broken down into very simple terms, that: A trab will get your pressure lower, at the cost of more complications. Right? So a tube — you might not get as low as a trab, even though statistically they might look very similar, from an IOP standpoint. You tend to have a slightly higher pressure with a tube. You’re going to get more complications early on from the trabeculectomy, for all of the reasons that we were discussing before. There is a push towards doing more primary tubes. So you don’t do the trab before doing the tube. And we’re just now starting to get data on that from Steve Getty out of Bascom Palmer, who is now presenting the primary tube study data, and it’s showing fairly good outcomes. So we have a lot of surgeons in the US who are going straight to tubes, and not doing trabs anymore. I’m surprised at how many surgeons in our circles, in the glaucoma circles, are actually not doing trabeculectomy at all anymore. They’re going from angle procedures, MIGS-type procedures — these are the microinvasive surgical procedures — straight into doing a primary Baerveldt or Ahmed. And I think that should tell you a little bit about some of the complications that we’re seeing from trabs, and how we’re trying avoid them at all costs. So I’m gonna ask one of my questions here, and I don’t know if we have the AR capability, the audience response capability. So I’ll ask this and see if we can get the results. See how many of you are paying attention. When comparing trab versus tube in the TVT trial, which of the following is true? Trabs are more likely to have complications. Tubes are more likely to have complications. Trabs and tubes are equally likely to have complications. Complications with filtration surgery would be highly dependent on surgeon experience. What’s the answer? What is it? How do I do that? How about we just raise hands? How many say it’s A? How many say it’s B? Come on. Somebody say it’s B. C? D? All right. You guys are all paying attention. All right. So that was easy. I’ll never be somebody writing test questions for any type of board exam. Trabs are more likely to have complications. So I’ll just summarize this piece, and then we’ll go to Hunter to do some questions. Filtration surgery is typically reserved for patients who have failed medications and laser therapy. I think that still holds true. We try to do everything prior to doing filtration surgery, because of the potential for complications. Some patients may benefit from a trab or tube without going through the typical treatment algorithm. We’re still in the art of medicine. We still have to look at the specific patient, and not just do this recipe, cookbook-type approach to things. Basic techniques and sound surgical management is important, but postoperative follow-up is equally important. The easy part, typically, is the surgery. Ex-PRESS device implantation may have some benefits over traditional trabeculectomy. We went over that slightly. The quicker vision recovery that can happen. Glaucoma drainage devices continue to evolve, and selecting the right device for the right patient is key. So Baerveldt versus Ahmed. I tend to pick the Baerveldts for patients who are the routine patients, where their pressure isn’t sky-high, and I’m trying to get it down to maybe the mid-teens. If the patient comes in at a pressure of 40, 50, 60, I do an Ahmed, because I know I don’t really have time to wait for that suture to open up. Those are the patients where I tend to use the Ahmed on more. And we continue to learn. We learn from each other, and certainly, over the past week, mingling with a lot of you, and being in the OR with a lot of you, I’ve learned — I feel like I’m gonna be a better surgeon after this week here in Barbados. So thank you very much.

May 25, 2018

Last Updated: October 31, 2022

Leave a Comment