This Live Lectures covers the basics of Glaucoma Drainage Device Surgery. Rationale for using glaucoma drainage devices, types of glaucoma drainage devices, indications, surgical techniques, comparative studies, and recommendations are also discussed.
Lecturer: Dr. James D. Brandt
DR BRANDT: So I’m gonna spend about — I don’t know. 30 minutes or so, 40 minutes, going through the basics of glaucoma drainage device surgery, and then when I’m out there, in a week or two, we’ll look forward to sharing with you some of the techniques that we use, particularly in children. And work with you. It sounds like we have a lot of cases to deal with when I’m out there, so I’m looking forward to getting to know all of you and working with you. I’m gonna go through a very basic talk, outlining the rationale. Let’s see. Here we go. The rationale for using these devices, the types of devices that are out there. The indications, the surgical techniques, the comparative studies, and then on to some recommendations, overall recommendations of when and where they should be used. And the rationale by which glaucoma drainage devices were developed was because trabeculectomies failed. And there was a great interest in the 1970s and on in developing something that would be doable after a trabeculectomy has failed. And trabeculectomies fail early because the internal ostium becomes closed by iris, vitreous, lens material, whatever other things that can physically close it, and more commonly, if the procedure was done properly, the procedures fail because of external fibrosis, either of the flap down to the sclera or scarring of the conjunctiva to the sclera. That process goes on for a long time, probably for years, and so trabeculectomies fail further on down the line for the same sorts of reasons of external fibrosis, but also when we use antimetabolites like mitomycin, the patients sometimes develop leaks and infections. And here’s an example of a young man I took care of. He was, I think, a teenager when we ended up operating on him. And he was lost to follow-up, disappeared, and then came back some ten years later, with a painful red and now nearly blind eye. And as you can see, he developed a late bleb leak and a bleb-related infection. Obviously this is not something that you want to see — especially on a Friday afternoon — and this underlies the reason that many of us have been working and trying to find alternatives to trabeculectomy, which, although it works, is not perfect, as we all know. The risk factors for failure of trabeculectomies include prior trabeculectomy failure. So if a patient has had a trabeculectomy in the superior nasal quadrant and it fails because of scarring, and you decide to go ahead and do a trabeculectomy in an adjacent quadrant, the fact that they failed once tells you that they are already primed to scar down a subsequent trabeculectomy. And certainly I’ve seen plenty of eyes that have had two, three, four trabeculectomies done, and each one fails faster than the previous one — presumably because you’ve primed all the fibroblasts to scar any insult to the eye. Any prior conjunctival surgery, such as previous large incision extracap and even small incision extracap, if the conjunctiva was violated, predisposes trabeculectomies to failure. And other things that are risk factors for trabeculectomy failure include younger patients, who have a more vigorous healing response, as well as doing trabeculectomies without the use of antimetabolites, which has become the standard of care — at least in this country — since the late ’80s, early ’90s. Risk factors for — and that’s a double-edged sword, the use of metabolites, because risk factors for late failure include the use of antimetabolites, which, as we know, lead to thin blebs, bleb leaks, and bleb infection. And on top of all that, we have the risk of — that’s included with poor hygiene and external eye disease. So a patient with bad lid hygiene, chronic blepharitis, is at big risk of developing a late onset bleb infection. So glaucoma drainage devices were designed and invented in the mid to late ’70s to overcome the two problems with trabeculectomy. One is to prevent the closure of the fistula that diverts aqueous out of the eye, and that is done with a silicone tube. And then the later failure that’s due to conjunctival scarring down to the episclera and sclera — the key innovation was to place an external plate that maintains that potential space with a material that conjunctiva cannot scar to. And allows the creation of a capsule that is hopefully permeable enough for fluid to get through it. People have tried sticking tubes in eyes since the early part of the last century. People have tried little glass tubes and other things like that. But we know that if you just put a tube into the anterior chamber and then into the subconjunctival space, you’ll end one hypotony, with flat chambers and all the other problems that result from that. And furthermore, even if you get away with no problems with hypotony, generally the conjunctiva just scars down very quickly after a few weeks or months. So the key innovation that was developed by Tony Molteno and Dr. Shocket — he’s in New Zealand. He’s in his 90s. Wonderful gentleman. And Stanley Shocket, who was a retinal surgeon in Baltimore — they both independently came up with an idea that you could create a tube that goes into the eye, and then diverts aqueous humor to a space maintained by this material to prevent scarring. And the Molteno implant was the first of these devices that was designed specifically for glaucoma. Dr. Shocket, a retina surgeon, came up with the idea to use in his patients with neovascular glaucoma. The idea he came up with was to take a tube and divert aqueous to the potential space created by an encircling scleral buckle. And I believe that he did this first in a few patients that had scleral buckles, and it worked, and so he essentially assembled a device made of a silicone tube from oculoplastic surgery, as well as retinal explant material, and assembled these and put them in. I did several of these in my fellowship in the late ’80s. They work well, but it’s a lot of surgery. Requires taking down the conjunctiva, 360 degrees. And is a big hassle to do. And I don’t think many people do them anymore. The two primary classes of glaucoma drainage devices that are used now are the non-valved implant and the valved implant. I believe that we’ll be working primarily with the Ahmed valve at KATH in a week or two. And these are generally my preferred implants in children, and we’ll talk more about that when I’m there. But I do really like the Baerveldt implant, which is shown in the drawing down below. This is a non-valved implant. And we have several comparative studies suggesting that the Baerveldt implant may result in lower intraocular pressures than the Ahmed does when you follow patients for a long time. The drawing that you see at the bottom left of the screen is actually not of the Baerveldt implant, but is in fact a drawing from the literature for the Addi, which is a complete copy of the Baerveldt implant, made by the Aravind hospitals in India. It’s not approved in the United States, but I’ve used it in a number of countries around the world with Orbis. And we can talk about this when I’m out there, but this is much, much less expensive, obviously, than the Baerveldt, but also much less expensive than the Ahmed, costing only about $50 or $60US, I believe, when purchased in packs of 10. So it is something that we can talk about, getting you guys some exposure to these devices. Because it’s probably much more cost effective in your hands there. The non-valved devices, as I said earlier, if you just stick a tube in the eye, the pressure will go to zero and you’ll get all those problems. So the non-valved devices require temporarily occluding the tube to prevent that early hypotony, and preferably for about 4 to 6 weeks to allow a capsule to form around the external device, or the external plate. I trained under George Baerveldt, who actually developed the Baerveldt implant — obviously the name — and he had learned his technique from Tony Molteno, so there’s a long pedigree here. But he taught me to ligate the tube with an absorbable suture. And I generally tie the tube off with 7 or 8-0 vicryl, which then dissolves after 4 or 6 weeks. Some people occlude the tube internally with a nylon suture, and then extend it under the conjunctiva, and they leave a long piece of suture that they just direct underneath the conjunctiva, to the inferior temporal quadrant, and then when it’s time for the suture to come out at the slit lamp, they make a little cut over the suture in the — 180 degrees or 90 degrees away from where the implant is, and they just pull that suture, which then allows flow to begin. They call that the rip-cord technique, like the rip-cord on a parachute. I’ve tried it a couple of times. I just find it cumbersome, and prefer to just tie off the tubes. Others have filled the tubes and the anterior chamber with a highly retentive viscoelastic, such as Healon 5, which can take weeks to clear. And essentially they temporarily occlude things just by gumming up the works with the viscoelastic. The potential advantages of the non-valved implant are, first of all, in the current designs of the Baerveldt implant, the double plate Molteno, and so on, are that they have a larger surface area for the explant or the plate that is placed. And there is some reasonable animal data that suggests that pressure lowering is roughly proportional to explant surface area. That said, there was a study done in humans that compared the 250 to the 350 to the 500 — I think it was 500 or 450 — square millimeter Baerveldt in — this was in the ’80s, early ’90s, that basically showed that, yes, IOP lowering is roughly proportional, but when you get too big, you increase the risk of complications, and so on. So most surgeons in the United States who do Baerveldt implants stay with the 250 or 350 implant. The big theoretical advantage of these implants are that the bleb is not exposed to the aqueous humor, which is filled with blood and inflammatory mediators, and so on. And so the thought is that the scar that forms around the plate externally is a much thinner and more permeable bleb than one that is formed after something like an Ahmed implant, where flow starts immediately, and during that early healing process, the bleb tissue is exposed to all sorts of inflammatory media that cause a more vigorous healing response. The Ahmed-Baerveldt comparison study, which I was involved in, and then the Ahmed versus Baerveldt study, which was a similar study done in Canada — we compared the Ahmeds to the Baerveldts. We now have five-year data. And it suggests that the Baerveldt has slightly higher early complications, but better long-term outcomes, in terms of pressure lowering. The Ahmed valve, which I think you guys are familiar with, uses a one-way silicone valve, and it’s actually not a true valve. It’s essentially a one-way flow restrictor. And they claim that it opens at 10 milligrams of mercury. On a bench, using balanced salt solution, yes. It does reliably open at 10 millimeters. But in vivo, in the actual human eye, it opens at variable pressures. As we all know, the pressures after these procedures can be anywhere from 2 to 25 or 30, depending on how viscous the aqueous humor is, with fibrin and so on. But the main advantage of this is that it works right away, and the valve and flow restriction, at least in most cases, avoids early postoperative hypotony. So the potential advantages of these devices are that there’s no need for temporarily occluding the device. You get immediate IOP lowering. You do have the advantage of single quadrant implantation, although the Baerveldt also — and I’ll demonstrate the technique there that can be used for either the Baerveldt or Ahmed implant — can be put in in one quadrant. It is slightly faster surgery than the Baerveldt, because you don’t have to lift up the muscles or anything like that. But most of these cases, with experience and an eye that doesn’t have a lot of scarring — these procedures, with experience, can be done in 30 to 40 minutes, if you have good exposure. The potential disadvantages is the converse of what we talked about for the Baerveldt implant, which is that the tissue that is gonna form the capsule over the plate is exposed to an inflamed aqueous, and the thought is that particularly transforming growth factor beta probably leads to a thicker and less permeable capsule. It also has a smaller surface area than the Baerveldt 350, and, again, that may ultimately restrict the degree of pressure lowering that we achieve. So the indications for these devices are failed trabeculectomies, extensive conjunctival scarring. Another one that many people don’t think about, that often goes into my decision making, is if the patient is anticipated to need further intraocular surgery. In other words, if a patient comes in that I’m sharing with the cornea service or the retina service, and they say — well, we need to get the pressure down, but then we’re gonna need to do a vitrectomy, or we’re gonna need to do a lensectomy vitrectomy, or we’re gonna need to do a penetrating keratoplasty… Those procedures, if you do a trabeculectomy and pat yourself on the back that you got everything under control — that subsequent surgery is probably going to cause a trabeculectomy to fail. And so we know from experience — not a lot of good hard data on this — but based on experience, glaucoma drainage devices tend to tolerate the stress and insult of further eye surgery much better than a trabeculectomy. So in a patient where I’m dealing with an accuse high pressure, but I know they are gonna need multiple other surgeries, I will definitely put a glaucoma drainage device in, even as the primary procedure, because I know that it’s gonna survive the patient’s subsequent surgical path more easily than a trabeculectomy will. And again — and we’ll get back to this — I believe that the glaucoma drainage devices are a very valid procedure to be doing as primary surgery in selected populations. And towards the end, we’ll talk about that. The techniques include conjunctival flaps and the preparation of the device. For the conjunctival flap, I was taught to do limbal-based conjunctival flaps. That means the incision is far in the fornix and the conjunctival flap is hinged at the limbus, as opposed to fornix-based, where a large limbal incision is made, encompassing 3 to 5 clock hours. And I think the limbal-based technique, which I use, is much better in children for a number of reasons, which we’ll talk about when I’m out there. But particularly in children with aniridia and aphakia, I much prefer the limbal-based technique. And I generally do the limbal-based technique in both children and adults. The implant needs to be prepared, so for the non-valved implants, I need to ligate it. I used to do it on the back table, while my resident or fellow got the eye prepped and set up and everything, but now with my presbyopia, I just do it under the microscope, on the operative field. For the valved implants, it’s absolutely crucial that the device be flushed with saline or balanced salt solution, because those silicone valve leaflets tend to stick together during manufacturing, and you have to unstick them to make sure that the valve is working, before you implant it. If you fail to do that, you’ve essentially put a tube in the eye that has a blocked-off end, and the aqueous has nowhere to go, and the procedure will fail. Then we trim the tube to an appropriate length. Once it is secured to the sclera. And we perform — we implant the tube into either the anterior chamber or, in an eye that has had a thorough vitrectomy, into the pars plana. But we do this under a scleral flap, under donor tissue, or the technique that I’m doing now is I make a very long scleral tunnel, using a crescent blade. And I’ll demonstrate that technique. I’ve found it very, very useful, especially in children. And we’ll talk about that. So here is my limbal-based procedure. This is a 4-week-old, who had bilateral congenital glaucoma of the newborn. We know that when congenital glaucoma shows up at birth, that’s a very poor prognosis. I generally will try angle surgery, but won’t wait very long to move on, because the success rate for angle surgery in a child presenting at birth is very, very low. Probably a good bit less than 50%. If I can get angle surgery to work, great. Because I’ve spared the conjunctiva. But I will move on very quickly to put a glaucoma drainage device in these children. Here is the Ahmed implant. I’m hoping you’re able to see the video reasonably well. And you can’t see off the screen that I’ve cannulated — and here you can see my forceps is pointing to a little bubble of fluid, and then it’s gonna pop the leaflets open. And I’m using quite a bit of pressure on the cannula, on the syringe, and now it’s popped through. You have flow through the valve, and now it’s safe to implant. Here is a Baerveldt implant. I’ve already made my — this is an adult. I’ve already made my conjunctival incision and cleared adhesions. And this is a technique that I’ve used and really can be done without an assistant. It’s easier to do with an assistant, but is very doable through about a 6 or 7-millimeter conjunctival incision. And then I switch hands, loop — this is the lateral rectus that I’m looping — and then I can tuck the device under the lateral rectus muscle. And then I pick things up, and here you can see it’s anchored under the muscle. Both superiorly and temporally. You don’t want to yank it right up against the muscle, because that’s a little bit more likely to cause a strabismus. So I usually anchor it in place a millimeter or two behind the muscle, so it’s not impinging right on the tendinous insertion of the muscles. Here is a nice clamp that I use, that is helpful for a nurse or an assistant, and this is a little technique with a little bit of dye there, to see exactly where it’s going in. I don’t use this technique much anymore, but it does have the advantage of showing you exactly where you entered the eye. Because sometimes, if there’s a lot of blood, it’s sometimes difficult to see where you entered. And I guess the video stopped before I could show you. Let’s see the next one. I guess not. So if we go back here… I’ve created an entry into the eye, just behind the limbus. And my assistant just flips the… I’ll go back here. Let me just go back and just show you. That’s here. So my assistant is holding there — and you can even do this yourself, or hold it up with a sponge — if you don’t have an assistant. This is a 23-gauge disposable needle, which is about the right size to fit the implant snugly. We sometimes use smaller needles, such as a 25-gauge needle in children, just because their sclera is more flexible, and you want a slightly tighter fit. But here the key is: My assistant is holding the conjunctiva over, so that I can see the needle as it enters the eye, and I want to make sure that I’m not going under the iris. It’s important in patients with angle closure, where there are synechiae, that you do everything you can to avoid going under the iris. In some cases, I’ll even do a surgical iridectomy in that area, and plan on having the tube come in a little bit posteriorly, but then go anteriorly in front of — through the PI, and in front of the iris. And I’ll show you guys that technique when I’m out there. So there’s a long history, as I said, of fornix-based conjunctival incisions. And most surgeons do the fornix-based incisions. In the tube versus trabeculectomy study, most of the US surgeons did fornix-based flaps, but both Tony Molteno and George Baerveldt usually implanted them under a limbus-based flap, and again, that was the way I was trained and I continue to do it. The advantages of doing these limbal-based is that you get better exposure posteriorly, it’s a much smaller incision, and you can make sure that it’s absolutely watertight. It tends to be more comfortable for the patient, because the incision is up in the fornix. And in children that are aphakes, that you’re gonna be working with contact lenses — you are quicker to fit them in contact lenses. And the big advantage in my mind, especially in aniridia patients, are that it spares the limbal stem cells. The disadvantages is that it requires a conjunctival clamp and a good assistant. And if there’s a lot of scarring, dissecting forward to where you’re gonna put the tube in can be a little challenging, and you risk buttonholing the conjunctiva. The main criticism of the technique is that it places an area of active wound healing right near the bleb. And many people moved to the fornix-based conjunctival incision because they thought this would cause these devices to fail more frequently. The advantages of the fornix-based flaps, which I’m assuming — but I don’t know this for sure — that you folks are doing — the advantages are that the limbal wound is far away from the posterior bleb. But the disadvantages are that they tend to be uncomfortable for patients, the wound is rarely completely watertight, and unless you make a pretty big incision, suturing the plate to the sclera can be a little challenging. And certainly in aniridics, when you do a 4 or 5 clock-hour incision at the limbus and cauterize in that area, you’re gonna destroy many clock hours of limbal stem cells, which the aniridics in particular need to keep their cornea clear. We did a study here at UC Davis, comparing my results to my partner’s results. She does fornix-based. I do limbal-based. And we found really no difference in the outcomes between these two. I can provide you with these papers, when I get out there. Again, in the interests of time, this is getting a little esoteric, so I won’t go into this. But both techniques work, and it really comes down to both surgeon preference and patient selection. And so if there’s no difference in the outcomes, why would you change what you’re already doing? And again, it is based primarily, I think, on the ocular surface. Here’s a patient with stem cell deficiency. You can see the staining, chronically, on their ocular surface. This is not an aniridic patient, but a patient that had a scleral buckle, so they had a complete peritomy, and then they had a trabeculectomy that failed. And so on and so forth. And they’ve destroyed — they have so limited a reserve of limbal stem cells that they have trouble epithelializing their cornea. And again, this is the disadvantage, I think, of the fornix-based flaps. And here’s a technique — here you’ll see a more recent video. This is my technique of limbal-based surgery. This is another kind of assistant clamp here. I’m putting in an Ahmed implant. And I tend to secure these to the sclera using 8-0 nylon. I was trained by Dr. Baerveldt to use a braided polypropylene suture. But I found that those end up eroding, years later. So I’ve moved to using the nylon suture, because it dissolves after a year or so. This is a mini-crescent blade, that I’ve found very useful. I’m gonna bring a box of them to demonstrate with and leave for you. But here you can see that I don’t need to make a flap. I don’t need to use any patch material. And here I’ve created an entry into the eye. Now, here is the Baerveldt implant, sort of duplication of what I showed you a few moments ago. And… We’ll zoom out here in a second. And with a little experience, you can actually do these quite nicely without an assistant. But it is very helpful to use these little clamps, which I’ll demonstrate to you when I’m out there, that hold the tissue very nicely. Here again I’m using the 8-0 nylon to secure the implant to the sclera. You’ll notice that there’s blood traveling down the tube, so I had not yet tied off the tube. But that’s not shown. Very straightforward. Now, this is a young woman who I’d been following since her teens. She’s in her 20s now. She has an aniridic cataract. So I did her cataract surgery. I pre-placed that implant and did those scleral tunnels. Now I’m going in, after having removed her cataract, and I’m sorry the video is a little more zoomed in than I’d like for demonstration purposes. But here you can see I’m entering her eye with that 23-gauge needle. And now I’m feeding the device into the eye. And you can see — I don’t know what the quality is of the video. Now, this is the closure, posteriorly. In her, I did place a patch graft of pericardium. But this is what she looked like afterwards, and her vision improved from around 20/400 up to about 20/60, which is where she is now. And notice that the conjunctiva at the limbus is all untouched. So we didn’t destroy all her limb bag stem cells and lead her to an epithelial failure and scarring. So, again, I generally do these limbus-based flaps in most of my patients. And this is what she looked like afterwards. You can see that she still has keratinization of her ocular surface, but the cataract was a big cause of her poor vision. And she has improved significantly. And for the very first time, I was able to see her nerve in many years, and happily, she still has some good optic nerve reserve. So she should maintain some vision for much of her life. As you probably know, in aniridics, as long as you can keep their pressure under control, what causes them to lose vision is generally their corneal failure, not glaucoma, as long as you maintain their pressures in a reasonable range. So I’ll skip through this discussion. Other than when you’re thinking about these patients, you should always examine the ocular surface, and recognize that any fornix-based surgery is gonna make these patients worse. I’ll move on here to… This is the postoperative care. I generally do subconjunctival injections of antibiotics and steroid. I’ll often put them on topical steroids pretty frequently for the first few days, and then four times daily continue the antibiotics for about a week. And then continue at least four times a day steroid for a couple of months. These patients — both the Ahmed — particularly the Ahmed implants, but also the Baerveldt implants — will develop what we call a hypertensive phase, and it usually occurs 4 to 6 weeks after there is flow through the system. So in an Ahmed implant, it’s 4 to 6 weeks after the surgery. In a Baerveldt implant or a non-valved implant, this hypertensive phase will often happen 4 to 6 weeks after the ligature releases and the tube opens up. And the key here is to treat these eyes with very aggressive aqueous suppression, to lower the pressure, and that allows a thinner capsule to develop. In terms of comparative studies, there are many case series in the literature, but the two major prospective randomized trials include the tube versus trabeculectomy study and the primary TVT study. This slide is wrong. I gave this talk in November/December. The primary TVT study reported its results at the Academy last November or October. And I’m one of the co-authors of the paper. We just submitted the manuscript. So you should probably see the primary TVT study in the journal Ophthalmology, within the next six months. But I’ll be happy to talk with you guys when I’m out there, on the results. But the tube versus trabeculectomy study… With which you’re probably familiar, and I’m happy to share all of these papers with you by email… Leonard, we should talk about — I can probably set up a repository on Cybersight, or send a Dropbox link for folks to download a bunch of these articles. The TVT study was done to compare a Baerveldt implant to a standard trabeculectomy in patients felt to be at high risk of failing a standard trabeculectomy. And we had 212 patients at our centers. And the inclusion criteria were these. They had to be adults. Less than 85. They had to have pressures between 18 and 40. On medications. And the other risk factors included a prior cataract surgery, trabeculectomy, or both. And these are the exclusion criteria. Again, I don’t need to go into these in great deal. We could have a mini-journal club when I’m out at KATH, and we can discuss some of these papers in more detail in person.. So we had a good retention of our patients over a course of 5 years. And you can see that ultimately, the intraocular pressures were about the same between the two. Not statistically significant after 1 year. But the failure rate was quite a bit higher for the trabeculectomy group than the Baerveldt implant group. And if you set different thresholds for what is considered a failure, that difference still occurred with the Baerveldt implants lasting longer than trabeculectomies. Now, often the critique of the TVT study is that — well, you’re doing this surgery to get people off of medications. And even at 1 year, most of the patients in the tube group were back on at least one medication. And that is true. However, if you follow patients long enough, as I’ll show you here, that difference went away after 3 to 5 years. So there was no statistical difference between the tube group and trabeculectomy group in terms of their medication burden after 3 years. So if you want a totally unqualified success, where the patient has a low pressure and doesn’t need to take medications, you are more likely, probably, to achieve that with a trabeculectomy for the first couple of years, but that advantage of trabeculectomy disappeared after about 2 years. So after 5 years, tube surgery was more likely than trabeculectomy to help you control pressure, avoid hypotony, avoid reoperations, and have bad outcomes like loss of light perception. But by the time you got out to 5 years, there was no real difference in terms of pressure reduction or the need for medical therapy. The complications were a little bit more common after trabeculectomy, but the bad complications, such as choroidal hemorrhages, and cataracts requiring subsequent surgery, and other things like that, were similar for both procedures. So my recommendation is that these devices are an important and proven option for surgical glaucoma management. If you haven’t done these devices, done these surgeries, it is a different skill set for the surgeon, but most skilled ophthalmic surgeons can learn this procedure. In fact, it’s my preferred procedure to take my residents through, if they’re gonna do glaucoma surgery. I think they get into fewer dangerous situations doing tubes than they do trabeculectomies. Trabeculectomy, in my mind, is more of an art, whereas these devices are more of a pretty standardized procedure. And at least in the TVT study, we found very similar results. And again, the second bullet point here, I think, is an important one, which is: Repeating trabeculectomies over and over again really just achieves limited returns with increasing complications. And I suspect that in your population, especially your adult population, you have a significant number of failed trabeculectomies, because of scarring, young patients, and limited follow-up, and just repeating trabeculectomies over and over again — I hope that I’ll be able to assist you in beefing up your skills and putting in Ahmed implants, and maybe getting a Baerveldt implant variant like this one, so that you can start using these. And I think personally that glaucoma drainage devices have significant advantages as a primary surgery when you’re anticipating that the patient’s follow-up is gonna be difficult. If they live many hours away, have limited resources to be able to come back from postop visits, I’d much rather put in a glaucoma drainage device. Because if it’s gonna work, it’s gonna work regardless of whether or not I saw them postoperatively, but as you all know, trabeculectomies need a lot of tender, loving care in the first month or two, to assure that they work. And if you’re not seeing the patient, that’s a recipe for the trabeculectomy to fail. So reemphasizing that, I recommend them as primary procedures when follow-up is anticipated to be long. Contact lenses-dependent. If the patients do a lot of water sports or are exposed to dirty water a lot, and also when pressure lowering is urgent, as we talked about. If they’re gonna need further surgery for other ocular issues, a tube is much more likely to survive subsequent surgery. The TVT results are very useful. Again, I’ll skip through this. This is going a little longer than I thought. But this really has lowered my practice for recommending these devices in many of these settings. And established ethically, the TVT did — the basis for the basis for the primary TVT, and just reported its results, and we’ll talk about the primary TVT results when I’m out there. Because I’m basically reviewing the final manuscript that’s being submitted next week. And I’ll try to put together a little slide set, so that I can give you a preview of the results, which basically showed that — no surprise — trabeculectomies are a little bit better than the Baerveldt implant in a primary surgery. But that’s only the one-year result. And I think what’s gonna be really interesting is to follow this group of patients out to 5 years. My suspicion is that many of the trabeculectomies will fail over the next few years, and will be in a similar situation where, long-term, there may not be a lot of difference. But that’s my gut feeling of what’s gonna happen with this group of 200 patients. But I don’t know that. Because we only have the 1-year results. So with that, I will close.