Lecture: Vitreomacular Surgery Update

In this lecture, Dr. Charles discusses indications for surgery including lamellar and partial thickness holes, epimacular membrane, VMT and macular schisis. Visualization and surgical techniques including staining and ILM peeling are also discussed.

Lecturer: Dr. Steve Charles, MD. Founder, Charles Retina Institute


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

DR CHARLES: Hi. This is Steve Charles in Memphis, Tennessee. I’ve just flown back from England, from the Royal College. And I’ll do the best I can to be articulate. I appreciate that 169 people so far have signed up for this seminar. Hopefully some more will join. I’m gonna talk this evening on vitreomacular surgery update. Obviously that encompasses a variety of diseases, and we’ll go through them one by one. One thing that I want to start out with is that, if you look at the top picture, you’ll see a typical epiretinal membrane. I usually show these at the opposite greyscale video, with a white tissue, if you will, on a black background. It shows better. But in this case, I inverted the video to make a case that numerous patients are said to have edema, based on central foveal thickness measurements, based on volume measurements, or sort of the topography-type map of thickness. This is simply not edema. It’s structural. And I call it foveal inversion. And it goes away. But it takes 18 months, when you peel the membrane. But I’ve had surgeons look at an image like this, postop, and say: Oh, they need intravitreal steroids. It’s edema. It’s not edema, there’s no cystic fluid spaces, and it will go away with time. Or I’ve had them misinterpret patients as saying… Oh, if you have cataract surgery, and you have an epimacular membrane, you’re more likely to get macular edema. No, you’re not any more likely it get inflammatory or Irvine gas macular edema than the next patient. It’s just thickness misinterpreted as edema. So now let’s move on to the impact of that. One is: Intravitreal steroids not only will not help, but there’s up to a 30% to 40% chance of steroid glaucoma, if you use triamcinolone, and virtually 100% of cataracts. So I would strongly encourage you to stay away from central foveal thickness except in a clinical trial context, or these topographic maps or color coding, and look at the black and white images, both preoperatively and postoperatively. I’ve also had another problem with respect to surgical indications, and that is: Patients like the image at the top have been said to have retinal disorganization. Not true. Or intraretinal fibrosis. Not true. Therefore, inoperable. Not true. The bottom picture is the same patient, after ILM peeling. But, as I’ll say many times, if you don’t ILM peel the situation, you will not get the level of structural improvement. The way I like to say it: If any tissue, vitreous, membrane, or even some other things, like hypotony, distort the macula, if you don’t peel off the ILM, it won’t go off to its normal shape. You must peel ILM to make the retina compliant, instead of elastic and taut, as it is when the membrane — the ILM’s in place. 27 or 25-gauge transconjunctival, sutureless. It’s the standard of care. There’s absolutely no reason to use 23-gauge. They have a higher incidence of wound leaks, et cetera. And I always use the forceps. So here I want to emphasize a point that I’ll emphasize over and over again: I am opposed to scrapers and picks and blades of any type. The notion that you must make an edge or look for the outside edge is simply incorrect. I’ve emphasized forceps membrane peeling now for 40 years, for four decades. I invented end-grasping forceps membrane peeling. But it needs to be done inside out, not starting with a scraper or a pick or a diamond-dusted scraper or any sharp needle to make the edge, but simply by pinch peeling. But that only works if the forceps grab right at the leading edge. More about that in upcoming slides. You can see that the forceps from Alcon and others have attempted to imitate this. I’m not here to sell Alcon products. Grasp right at the leading edge of the forceps. Not along some plate. So I don’t approve of the so-called asymmetric forceps, or of the alligator configuration forceps. I only approve of these that are smooth and grasp right at the leading edge. And this makes a major difference. It’s the enabling technology. At some point, I began calling this pinch peeling. This is another way to think about it. So end-grasping forceps membrane peeling is the standard, whether it’s the epimacular membrane, a macular hole for ILM peeling, vitreomacular schisis, vitreomacular traction syndrome, or peeling for other indications we’ll talk about. So what are the benefits? One, you don’t have a sharp instrument in the eye. And patients move on the operating table. These are done under local. And two, you don’t have the issue of trying to find the diaphanous thinner, outer margin of these so-called glassy or cellophane epiretinal membranes. If they’re thick, you can kind of see the outside edge. But if they’re thin and have specular reflection, kind of shiny, it’s very difficult to see the outside edge. So you don’t have to put a pointed object or something that damages the nerve fiber layer or with a lot of downforce on the retina, like some of the scrapers do. Or all of them do, actually. So, again, we want to start where? We want to start at the epicenter. What is that? How do I know what’s an epicenter? The stria and the retina point to the epicenter. That’s one way to tell. It’s somewhat more matte finished, rather than transparent. And it’s at the apparent center. Staining certainly helps, but I don’t use the term inverse staining. But some do, to turn out that when you stain with Brilliant Blue, which stains the internal limiting membrane, and is very safe, unlike ICG, which I do not use, although many are forced to use it for regulatory reasons. When you stain with those, if there’s membrane in place, then there’s no staining where the membrane’s still there, and that’s what some people call inverse staining. And that’s another useful technique. I never, ever use triamcinolone to mark the epiretinal membranes. It is a mistake, in my view. It is no more specific to the ILM than it is to seeing if there’s dust on your countertop at the hotel, to see if the maid had dusted it off. It is certainly useful to visualize vitreous, but now, when you use Ingenuity, you see vitreous so well anyhow. I really find no need for triamcinolone, particularly in marking the vitreous, and I’ve never seen any need to use triamcinolone for the ILM. So what are my indications for ILM peeling? All lamellar and full thickness holes. Those who, for a variety of reasons, somehow feel that ILM peeling is dangerous or traumatic, say — well, it’s only a partial thickness hole. I don’t need to ILM peel. I disagree with that. If you want to return a partial thickness hole to a proper anatomy, foveal anatomy, to restore normal anatomic configuration, it’s not gonna happen if you don’t peel the ILM. So what I’ve seen over and over again — people say oh, it’s a partial thickness hole. It’s symptomatic, therefore I’m gonna do a vitrectomy and peel off the PVD. Many of these patients already have a PVD with the lamellar macular holes. But peeling vitreous away from the optic nerve head allows gas bubble access to the retinal surface, but it has nothing to do with fixing the macular hole per se. The gas bubble does, for reasons we’ll discuss. So I use ILM peeling in every epimacular membrane case, every vitreomacular schisis case, all vitreomacular traction syndrome, whether they’re diabetic or not. Many diabetic traction detachments, and frankly, more and more PVR cases. I learned from a Spanish author that hypotony maculopathy can be treated, and I haven’t had a large group of cases from the glaucoma specialists, but I have seen some, and they actually do quite well. If you do internal gradients of subretinal fluid and internal fluid-air exchange improperly, when fixing rhegmatogenous retinal detachments, you can create a macular fold. This usually occurs because you do fluid-air exchange before drainage, and you push retina posteriorly. And those folds — I do not find any need, if someone creates a fold, to redetach the retina. If you simply peel ILM, over time the retina will restore to near-normal configuration. So what are the advantages of ILM peeling in the context of epimacular membrane surgery? One is it reduces the recurrence rates, as Anselm Kampnik from Germany pointed out. This was my wake-up call, when Professor Kampnik brought this to our attention. Why is that the case? Because the basement membrane for what we at the time thought were modified astrocytes that caused glial recurrences — we now know them to be Mueller cell — are the progenitor cells for reactive gliosis. Whether it’s PPR or diabetic recurrences on the retinal surface, or epimacular membrane recurrences, it’s Mueller cell reactive gliosis that’s occurring. At any rate, if you eliminate the basement membrane, then you eliminate this problem, for the most part. We have an incredibly low recurrence rate. Probably if you really have a good view, the patient holds still on the table, it’s essentially zero, once you ILM peel. Secondly, if you don’t ILM peel, the striae will not go away. The retina is sort of fixed in that position by the ILM, in that folded position. So I had the experience over many years, before ILM peeling was introduced for this application, of peeling off a membrane, documenting it on video, and then postoperatively, the folds were still there. And the referring doctor said: Gee, didn’t you peel the membrane? So I’ve learned my lesson. When the striae are completely gone, there’s still room for further visual improvement, lasting out to at least 18 months after surgery. Why is that? Because rods and cones are shaped like rods and cones, and when light is on-axis, they are 30 dBs more sensitive than they are when it’s off-axis. This is called the Stiles Crawford effect. There’s not a clinical machine for testing that. But J E Noch at Ohio State did a lot of work in this area, and this is clearly the case. And so fortunately photoreceptors adaptively realign at the nodal point. But it takes many, many months, and long after the retina appears normal, on OCT. So doctors that don’t check the patient’s vision 18 months or 2 years out have an incorrect impression about the level of visual improvement. I’m absolutely distraught if the patient isn’t 20/30 or better after epimacular membrane surgery, yet there’s papers in the literature that believe that 20/80 or 20/70 is an average outcome. That’s simply because ILM peeling was not properly done, or my fear is that there are many, many cases of ICG toxicity out there. The third reason to peel off the ILM is that you guarantee removal of everything anterior to it. Clearly if the ILM is gone, all epiretinal membrane and all residual vitreous cortex is gone. As we’ve learned over the years, seeing a Weiss ring in no way indicates that there’s a problem with — seeing a Weiss ring in no way indicates that a PVD has occurred, in the sense of complete removal of vitreous from the retinal surface. Vitreous remains on the retinal surface — I remind you, it’s the vitreous cortex, not the vitreous face. So there are multiple layers. This demonstrates the intimate relationship of the ILM and the Mueller cells. The Mueller cells of course are columnar. When you successfully peel ILM, on occasion you see what two of my ex-fellows — Eric Sigler and John Randolph reported — which is what they call intraretinal layer cystic change. This is not edema, and it doesn’t respond to steroids. It may respond somewhat to anti-VEGF compounds, we’re learning lately. But again, it’s not edemas. Some people call it retinal dimples, which I think is a ridiculous term. Others call it dissociated optic nerve fiber layer, which is not the problem either. The problem is avulsion of the Mueller cell foot plates, and that’s, in my view, why the term intraretinal area cystic change is more appropriate. How can you identify this on OCT? The little cystic spaces are columnar, following the shape of Mueller cells. And so there are certainly people who are overly aggressive with ILM peeling, using scrapers and so on and so forth, that actually do dissociate the nerve fiber layer. But if you use proper visualization and proper forceps techniques and don’t use scrapers, you will not routinely dissociate the optic nerve fiber, yet you will have this intraretinal area cystic change. It can actually be seen preoperatively, when there’s extremely aggressive traction as well. ILM staining is absolutely essential. I am stunned with how much I missed before ILM peeling was introduced. I use this in every single — I see staining in every single case of macular surgery, of that list that I illustrated earlier. Fortunately it’s approved in Europe and available in many other countries. You must be very careful if a compounding pharmacy makes this that it’s made properly. We know there was some off-brand PFO in Spain that ruined a number of eyes. This certainly could happen from Brilliant Blue as well. Although I’m not aware of much of this happening. Just a few isolated cases from bad compounding pharmacies. In the United States, it’s not FDA approved. So as I said, it has to be prepared by a compounding pharmacy. But fortunately it’s approved in Europe as Brilliant Peel, and available in Asia and many countries as well. What are the advantages of Brilliant Blue over ICG? It’s dissolved in BSS, not water. It’s not fluorescent, so there’s less risk of phototoxicity. It has a stable pH and osmolarity. There’s no need to infuse it under air. And one of the big advantages, in addition to not being toxic, is that you can repeat it. So it’s very common for me to inject Brilliant Blue two to three, four, five times in a case. I keep it in the refrigerator until immediately before use, so it sinks, from convection. So you don’t have to inject aggressively to get it on the retinal surface. And I use a 3-CC syringe, instead of a 1-CC syringe. This decreases the throw of my finger on the plunger, but more importantly, it slows down the jet that comes out the other end, and prevents you from making a little retinal break with hydraulic pressure, and causing Brilliant Blue to go underneath the retina. How should you view the macula? I am absolutely opposed to the use of the Resight or the BIOM for macular surgery. A plano contact lens gets rid of all corneal asphericity, think it’s absolutely an appropriate way to view the macula. It does not take another physician or surgeon to hold the contact lens. Technicians do it very, very well. Just tell them to keep it parallel to the floor, and don’t press down. And be nice to them. Don’t yell at them. And you’ll find that this non-contact viewing is far better for macular surgery, in my view. I’ve used Ingenuity for all cases. I know many don’t have access to it yet. This increases the depth of field by 2.6x, and is another tremendous advantage in doing macular surgery. The motion should be, first of all… This shows how you’re ingrasping in the epicenter of the membrane, not at some manufactured edge, or at the outer edge, outer margin. So again, epicenter. In the middle, where the striae point. And you pinch peel right at the — just at the level of the membrane. If you lift, and there’s the slightest distortion of the retina, let go and move to another location. With 27 and 25-gauge forceps, you can see them bend, and you interpret that as tactile feedback. There’s also some tactile feedback to your fingers, and you can tell the ILM, because it feels elastic, unlike retina. And the motion should be circular, not unlike a capsulorrhexis, and you can steer the direction that you turn. And this motion is the same, whether you’re peeling an epiretinal membrane or peeling ILM. The technique is the same. Sometimes the membrane is very adherent, and you’ll peel both in one layer. I make no attempt to try to be in one layer or be in two layers. I just grasp the surface at the epicenter and peel. And then I restain, and if the ILM went with it, the stain will be on the bottom of the epiretinal membrane in some areas. But when you restain, you’ll see that the ILM is gone. It’s commonly spoken on the podium, meeting after meeting, and it’s published in numerous places that vitrectomy causes cataract. It’s simply false. It’s been false forever, but it’s stated by surgeon after surgeon who wants to justify scleral buckling, when I’ve traded off scleral buckling for vitrectomy to repair all retinal detachments. But what does occur is inevitable progression of nuclear sclerotic cataract. If there’s no pre-op nuclear sclerosis, you don’t get a cataract from vitrectomy. I have had literally a thousand patients — estimated, of course — that have had vitrectomy surgery 10, 15, 20 years ago, maybe more, and they don’t have a cataract. Why? They were 20, 30, maybe 40, and had no cataract when we started. But if there’s any nuclear sclerosis present, it will progress. Why? It’s been shown by Nancy Holekamp, who actually measured it, and Einar Stefansson, who measured oxygen levels in patients, as well as Stanley Chang, who postulated that the oxygen tension in the vitreous cavity goes up by 12 millimeters of mercury — the partial pressure of oxygen goes up by 12 millimeters of mercury permanently. It has nothing to do with the infusion fluid. The infusion fluid, if you use offbrand infusion fluid, and you’re unwise enough to put dextrose in it, or to put bicarbonate in it, instead of using BSS or BSS Plus, or if you put epinephrine in it, you can cause a posterior subcapsular cataract, which is just simply chemical toxicity. I never add anything to the BSS Plus. I think it’s a big mistake. And I never see posterior subcapsular cataract during vitrectomy. I haven’t for 35 years, since BSS Plus was introduced. Numerous surgeons use this justification or the justification of only having to see the patient once to do phaco-vit in all their cases. If you’re a really good phaco surgeon, and numerous vitreoretinal surgeons in my view are not, yes, you can take the cataract out and put an implant in, and go on and do retinal surgery. But it’s not infrequent to have the pupil come down, or to have a little haze at the cornea, or a slight bit of striae, and now your visualization is compromised. But even if you’re perfect in your phaco, and there’s no compromise to your visualization, you have another issue. And that is: It’s very hard to get the right refractive outcomes when you’re doing combined procedures. Patients today prefer to be emmetropic after surgery, or to get their desired monovision or whatever it is. And if you’re using multifocal lenses or torics, which many patients want — I’m opposed to multifocal lenses in any patient with macular disease — but torics are a wonderful lens to use, and I know numerous vitreoretinal surgeons that routinely do phaco-vit when they shouldn’t, and they never use a toric. So now they’ve elected to not have a wonderful and safe treatment for the patient’s astigmatism, which I think is a mistake. So I strongly prefer separate procedures. So how do I decide? If the patient has a posterior subcap or has 3 or 4 posterior scleroses, the phaco surgeon does it first. We wait a month. And then I do the macular surgery. If the patient has one, maybe 2 posterior scleroses, I do the vitrectomy first, and then I send the patient to the cataract surgeon for cataract surgery. If they’re gonna do the surgery first, you must caution the cataract surgeon to measure the axial length optically with low coherence interferometry from the pigmented epithelium. So that means the Haag-Streit Lenstar 9000, or the Humphrey IOLMaster. You have to use optical means to measure axial length. If there’s substantial cataract, of course, you can’t do that. And that’s another issue. So let’s shift gears to talk about macular holes now. We’ve talked about membranes. What causes macular holes? And the answer is: Nobody knows. Yes, Don Gass put a theoretical explanation, which everybody feels compelled to describe in every single talk on macular holes, but it hasn’t been proven. For example, if 70% of the population get a PVD, what’s different about those patients that go on and get a macular hole? Why is their vitreous still on the retinal surface in some of the patients, and other patients, there’s not? Why is it 2 to 3 times as common in women as it is in men? Why does it wait until average age 59? So the reality is that we have no idea precisely why patients get macular holes. We know for certain that a Weiss ring does not mean there’s not vitreous on the retinal surface. The classification system is simply useless. It is erroneous. There’s numerous evidence of this in the literature, where people say… Oh, it’s a stage I, stage II, stage III, stage IV. You don’t know until you get to the operating room whether there’s vitreous or membrane in the retinal surface or not. Numerous patients — with ultrasound, it’s hopeless. Even with OCT, it’s hopeless to classify these patients. It is a mindless exercise that people go through. We know from Neil Kelly’s and Rob Windell’s pioneering work that macular hole surgery clearly did not — macular holes do not come about because there’s a full thickness piece of retina missing. Well, that was widely known to be the truth. Dick Green and others published that. Except it’s simply not true. There’s remarkable — as Mark Blumenkranz says — plasticity in return of near-normal foveal anatomy, where, only in a rare patient, can you even see the seam where the macula comes back together. So clearly there’s a plasticity role here, which is exciting. What about partial thickness macular hole surgery? I do not do that to prevent full thickness holes. One can simply not predict who’s gonna stay partial thickness, and we have a very effective treatment for full thickness macular holes. So I treat for one of two reasons: I treat partial thickness holes if they’re symptomatic, if the patients have metamorphopsia, they have distortion or decreased vision, or the vision seems dimmer in that eye, or defocused, which certainly can be the case with partial thickness holes. Lamellar holes. Or if there are structural changes visible on OCT that mean retinal damage is occurring. Schisis, subretinal fluid, or intraretinal fluid — that doesn’t go away by itself, and that doesn’t bode well for the long term health of the macula. So in those cases, I operate based on the findings, not necessarily the symptoms of metamorphopsia. But again, not to prevent progression to full thickness holes. You certainly accomplish that, but once again, you don’t know who to operate on and who not to operate on for that reason. But I emphasize, as I alluded to earlier, the need to do ILM peeling and use gas. Numerous people — oh, it’s a partial thickness hole. We don’t need to do gas. I promise you, you’re not gonna get satisfactory return to normal foveal anatomy without ILM peeling and without SF6. Well, let’s just use air. I don’t think air is as effective. Can I prove that to you statistically? No. Because I’m uncomfortable randomizing. When I’ve seen patients that others operate with air, I believe there’s a lower success rate. I have not carried that argument to the point of using C3F8, however. So I think that’s important. Now, if you look at this top picture, there’s a lamellar macular hole, some would call it. But there’s also macular schisis. I think there’s enough structural change there that this patient needed surgery. So I peel the ILM and use gas, and you can see near normalization. There’s very slight foveal eversion. This will go away by the 18-month mark. How do macular holes close? In my view, during phase I, what the ILM peel does is a couple of things. One is it makes sure there’s no posterior vitreous cortex or epiretinal membrane traction. But more importantly, it increases retinal compliance. And I’ll discuss that in a moment. And what does the gas do? The gas works by surface tension, not some dumb term, “tamponade”. “Tamponade” comes from the French. It means to plug, not to seal. So this isn’t just to keep air from going through the hole. Surface tension is an inward-directed force. We’ll talk more about that in a moment. But there’s additional benefits of having gas in the eye. So buoyancy, Archimedes’s principle, tells you where it goes, but not what it does. Tamponade, again, is not a physics term. So what does it do? One, the most important factor is this lateral surface tension effect, as Vinnie Reppucci reported at the Club Jules Gonin in Cape Town. Very, very central concept. But it also prevents trans-hole flow, and as I pointed out, it also prevents trans-retinal flow. Then there is, after apposition is achieved from the gas bubble, then there’s some healing, possibly from astrocytes, or more likely, now we know, from Mueller cells. I should have reworded the slide. Wollensak did an interesting experiment. Took rabbits and put the retina in a strain gauge, force measurement situation, and measured the elasticity of the retina. Measured its biomechanical properties. He then ablated just the ILM with the Excimer laser, and showed that the retina was 27% longer and there was 53.6% reduced force, so in short, you convert a tight retina to a limp, compliant retina, and that’s how the gas bubble is able to cause edge to edge joining of the retina. Of the edges of the retinal break. Let’s go back here, and let me talk a little bit about surface tension. Einstein’s first paper, when he was in the patent office, was actually on surface tension. If there was a Starbucks in those days, he would have probably figured it out that way. But he noted that if particles were present on the surface of a liquid, that they clumped. And he said… Now, wait a minute. How does a particle, you know, 100 microns away from another particle, know to clump? What force causes them to all move centrally and clump? And he could have said: Well, when kids have this little loop, and they blow bubbles, they don’t come out as a disc. They come out as a spherical bubble. When water droplets come out a faucet, they don’t come out as cylinders. They come out as droplets, with a little tail on them, because as they descend under the force of gravity, the air flow pushes them into a droplet. But they would otherwise be spherical, if it was a zero gravity situation. So that force, surface tension, is what closes the hole, as, again, Vinnie Reppucci pointed out. It pulls the retinal edges together, and ILM elasticity is what helps make that possible. We’ve talked about staining already, but here’s an example of ICG toxicity from a major group in Nashville, Tennessee. This patient had 20/400 vision from this massive ICG toxicity. The best way to see ICG toxicity is with the fundus autofluorescence, using the blue laser on the Heidelberg. I’ve said earlier that there are those who stated on the podium — I think Gujarav Shah is one of them — that triamcinolone is specific for the ILM. This is simply not the case. But in fact, anecdotal evidence indicates it reduces the closure rate of macular holes, and people have reported triamcinolone getting trapped in the hole, or actually entrained or trapped in the subretinal space. And again, it’s not a stain. So it can be used to see the vitreous. I find Ingenuity a far better way to do it. But long before I had Ingenuity, I never, ever used triamcinolone to mark the vitreous, before or during vitreous surgery. And this is particularly important in macular hole surgery, for the reasons I just stated. What’s the rationale for gas? In addition to all those points I made earlier about the surface tension effect, and this drying effect that I mentioned, probably the drying effect stimulates Mueller cells — again, stated as astrocytes here — to close the hole. What do you do if there’s a very large hole, and it won’t close? Well, the technique I did for several years, which I reported, is called arcuate retinotomy. I no longer use this, and I’ll tell you why in a moment. But Aneesh Neekhra was doing a research fellowship with Barry Kuperman at UCI, in California, and called me and said: Could you cut the retina somehow if there’s a large hole? So I made some drawings and decided that an arcuate cut, sort of parallel to the macular hole’s temporal edge, might be the best way to do it. And we reported a series of, I guess, over 50 cases — I don’t remember the number right now — that had roughly a 2/3 or a 66% success rate. And so we’ll show you what that theoretically looks like. And there’s some very, very nice cases. Again, 2/3 of the cases close. And there is quite good visual recovery. I’ve had edema in one case. I had retinal detachment in another case. But for the most part… And here’s an example. You can see the arcuate retinotomy temporally. But in this patient, the hole was present a long time, and although the hole closed, it’s very thin, and there’s, as you can see here, some foveal atrophy, so the vision was only 20/200. These are the scissors we used, the same ones used for diabetic traction attachments. But the game change some years ago — again, I apologize. This slide is out of date. There’s now 28 cases that I’ve done. Tamer Mahmoud, while at Duke — he’s now in Michigan — reported cutting a graft from peripheral retina and placing this in a macular hole, and doing direct PFO-silicone oil exchange to literally put neural tissue in the hole. I am not a fan of ILM flaps. I don’t do ILM flaps. It’s not neural tissue. But I am incredibly impressed with Tamer Mahmoud’s technique. I’ve made a number of changes in the technique, based on my own experience with it. There’s now in IRB-approved cases from five of us — Dr. Mahmoud’s series and my series — where we’ve now reported 49 cases, with a 90% anatomic success rate. So what do I do differently now? I’ll show what you this looks like. Here’s a beautiful graft. This is several months out. Silicone oil still in the eye. I have a 20/25, a 20/40, a 20/80, a 20/100, and a 20/200, and the average hole size that I operated on with this technique was 1,018 microns. 1,018 microns. All of them failed prior surgery with ILM peeling. ILM peeling was attempted again with Brilliant Blue staining on every one of these patients, when I went back in to do this. So I haven’t done any primary surgery case until a week ago, when I did a diabetic that had bilateral 1.5 disc diameter holes. Almost 2-millimeter holes. And I grafted one eye. So we’ll see how that works. But every other case had prior surgery. Sometimes the graft, when they’re first seen postop, is a little misshapen, as you see here. And then the shape alters with time, indicating some form of plasticity. I initially did the donor side superiorly. Because it was easier for my hand access. But now I make it just inferior to the inferotemporal vessels, because there’s gonna be a small scotoma at the donor site, and I thought it would be better if it was above fixation. It wouldn’t bother the patient. Here’s another graft, showing you the donor site, superiorly. In this patient, I operated twice. You can see two donor sites. But again, the graft tissue is in place, and even though the tissue looks funny, this patient had 20/60 vision! Here’s another pre-op patient that we fixed. This one, although I did a macular patch graft, I think this had primary closure, so we finally excluded this from the series. And I just wanted to be honest about that point. Another case. Again, the donor site superiorly at the graft, placed in the macula with improved vision. So the other change that I made in the series — we’ll go back one slide — is that instead of using silicone oil… Silicone oil has very low oxygen carrying capacity and a low extraction coefficient. Whereas liquid perfluorocarbon was designed as a blood substitute for Jehovah’s Witnesses, as well as for liquid ventilation in cases of chemical pneumonitis and acute respiratory failure. And so it was developed for high oxygen carrying capacity. So I now use medium term PFO. I’ve done over a thousand patients with retinal detachments, inferior, temporal, and nasal breaks, with PFO in the eye for two weeks. I have a 22-year experience with leaving PFO in the eye, and I’m extremely happen with it. It’s clearly nontoxic. If you use the Alcon Perfluoron. So I said: Let’s use that technique in these patients. So now I put the graft in place under PFO. I cut the graft with the scissors after ballooning up, making a bleb with the 38-gauge cannula and balanced salt. So I balloon up a bleb, I take the diathermy, treat just the vessels, then I cut it with scissors, and it’s still in place. Then I do a PFO injection with a MedOne dual bore cannula. So now we’ve got the PFO in place, the graft has been cut, and I take the forceps and grab the graft and drag it a little bit of the time into the macula, under the PFO, with my forceps maintained right at the retinal surface, so the graft stays unfolded and properly oriented, right side up. And slide this into the macula. And in one week, I go back and take the PFO out. So I’m very happy with this idea. What about… Let’s shift gears now and talk about vitreomacular traction syndrome. I don’t have a slide on it, but let me say from the outset: I am 100% opposed to the use of Jetrea or ocriplasmin. It was one of the questions presubmitted to me. One of the other questions submitted to me: What do I think about ILM flaps? And the answer is: I don’t do inverted ILM flaps, and I don’t do Jetrea, and I’ve never done Jetrea once. If there’s a patient that has significant vitreomacular traction, and for some reason does not have… We cannot take them to the operating room… Say they had severe kyphoscoliosis, or severe emphysema or congestive failure, where they couldn’t lay down on the operating table, or possibly an economic situation, where surgery couldn’t be afforded. Then pneumatic retinopexy, as reported by Clement Chang and Kelvin Mein and others, and Jorge Arroyo, is actually a far better technique. It’s safer, it’s essentially free, and it’s effective. Why not Jetrea? Jetrea, 1.5% of the time, has profound visual loss. And it has a very low success rate in the first place. 25%, maybe. Maybe 30%. But 1.5% of patients have profound visual loss, presumably from dissolving into the photoreceptor matrix. Some patients have dislocated lenses, subluxated lenses. Other patients will get a low level retinal detachment. The ERG is non-recordable in some. And in some of these cases, these complications do not go away, and you converted a patient with 20/30, 20/40 vision, preoperatively, with a so-called minimally invasive technique of injecting this ocriplasmin, to a patient that can’t see. And that’s simply unacceptable. So I do not use Jetrea in any instance. This patient is a very important patient. And I have numerous other slides like this. This patient was seen by numerous retinal experts who opined that this is some form of cystic degeneration of the macula, and nothing could be done. I said no. This is vitreomacular schisis. I peeled the ILM. This patient went from 20/200 to 20/25. The patient had it in both eyes, was 30 years old, was presented at major rounds in New York, as well as New Orleans, and all the experts said: You can’t do anything. And there are numerous other cases that I have of vitreomacular schisis that have been judged to be somehow inoperable, because of the cystic change. Not only does this structurally improve — more importantly, visual function improves dramatically. What I find fascinating, and I don’t really have insight into this, except to say it’s another example of biologic complexity, is that the genesis of vitreomacular disorders has something to do with posterior vitreous separation. Clearly we see many, many patients, as retinal specialists, where this is an acute event. Flashes, floaters, 5% incidence of retinal breaks. 6 out of 10,000 go on to get retinal detachment. Okay. So that’s the classic… But there are many, many patients that have a slowly evolving PVD. I had it in both eyes myself, and had a tiny little array of dots before the fovea. You could see them on OCT. Which corresponded to what I saw visually. And over the course of close to a year, they went completely away, as the vitreous finally peeled away from the retina. This is said to be due to glucose metabolism. I’m blocking on the right name for it right now. But at any rate, PVD is not always an acute, sudden event with flashes and floaters. It can be slow. But it is clearly the cause of epimacular membranes, as Roth and Fuss showed many years ago. So an ILM defect is created. As I said, we used to say it was modified astrocytes that caused the problem. We now know it’s reactive gliosis from Mueller cells. So glial repair makes epimacular membranes. We know that it has something to do with macular holes, but when there’s still vitreous on the retinal surface, exactly why does that happen? And then we had these other patients that, when the vitreous peels away, it remains secured to the fovea, presumably from some hypocellular contraction of the vitreous collagen matrix, and they get vitreomacular traction syndrome. Why one patient gets one and one patient gets the other is anybody’s guess. And again, there’s no genetic factors here, other than myopia, myopic vitreomacular schisis, and there’s no systemic factors either. Again, very complicated. So I find this fascinating, the fact that we have slow PVDs that are related to glycosylation end products — is what I meant to say earlier. So another story: There are some people in Vienna, Austria, and in Detroit, who seem to believe that vitreomacular traction syndrome is one of the causes of exacerbation of wet macular degeneration. I couldn’t disagree more. I think this is ridiculous. And frankly, I think part of it was self-serving by those who wanted to push Jetrea. But at any rate, what’s the reality? Vitreomacular traction syndrome is very, very common. Wet macular degeneration or neovascular AMD is also very common. And they both cause subretinal fluid, and they coexist. There’s absolutely no scientific evidence that they’re related. They both must be treated, and they’re treated separately. But sequencing matters. So I think it’s important to do the best you can to dry out the macula with anti-VEGF compounds such as Eylea, Lucentis, and Avastin first. At least two or three injections, and get as much drying as people say, as you can. And then do the vitrectomy to treat the vitreomacular traction component. If you do it first, you change the pharmacokinetics, so you shorten the duration. Shelley Keatts has done some good work in this area. So we shorten it by, I guess, two weeks, as I recall his presentation. And so I recommend treating with anti-VEGF compounds first. But again, these are unrelated disorders that just happen to coexist, because they’re both common. That’s my last slide. We had a tremendous number of questions that were submitted. And now if we can put those questions up, I’ll be glad to do what I can to answer them. So why don’t I believe in flow-based machines in vitreoretinal surgery? The primary reason is that I’ve tested them, in the lab, blinded, between the best of the peristaltic pump machines and the venturi-based machines, and I found that there’s a 250-millisecond advantage in terms of responsiveness when operating under air. Peristaltic pump machines simply don’t work under air. You have to have very precise control of vacuum at low levels, which the Constellation does, but there’s no proven advantage of the flow controllers. That said, the next generation of machines we’re working on at Alcon will be able to have flow control, but again, you cannot do vitrectomy under air, which is a very useful technique I’ve advocated for many, many years — you can’t do that at all with a peristaltic pump machine. I alluded to this in my talk. Inverted flap. I have not used the inverted flap, so I’m clearly not an expert. Clearly some very excellent surgeons have advocated this and used it. However, I moved to the full thickness macular patch graft that Tamer Mahmoud developed. So I don’t really have any reason to use an inverted flap. I want to put neural tissue in there, not basement membrane. Next question: Do I have any tips for avoiding creating a full thickness macular hole when I peel the ILM and there’s a large cyst? One, proper visualization is essential. Use the flat contact lens, not the BIOM or any other Resight. Number two: You must stain with Brilliant Blue. Or I suppose ICG, if you don’t have Brilliant Blue. I have no experience of ICG. Thirdly, you must peel toward the macula. And don’t lift. Peel tangential to the retinal surface, or kind of parallel to the retinal surface, and peel toward the macula. Never peel — never lift and never peel away from it. But some people actually will peel around and leave a tiny bit of ILM present, right over the macular cyst. I don’t have a problem with that. I have not intraoperatively created a macular hole. I’ve had a couple that occurred months down the line, but I haven’t had one that occurred intraoperatively. And now we change gears dramatically, and we’re talking about macular traction retinal detachment cases. I’ve developed scissor segmentation delamination, and it’s becoming a lost art, unfortunately. You cannot do safe, highly effective macular traction detachment surgery for diabetic traction detachment cases without having the capability of using curved scissors delamination. The vitreous cutters, with better fluidics, with the beveled cutting tip, with higher cutting rates certainly can do more than ever before, but if you believe you can do the entire case of a diabetic traction detachment using the vitreous cutter, you’re sadly wrong, and you’re gonna make a lot of retinal breaks. Day in, day out, people call me up and say: I’m doing a tough macular traction case. I’m planning to use membrane peeling. What? Membrane peeling? You don’t use membrane peeling with traction detachments. You must shear the membrane away from the retinal surface, either with the cutter or with the scissors. And I’m planning to use oil. Oil? Why would you use oil in a diabetic traction detachment? The vast majority of the oil cases are because you didn’t do the surgery right. You made retinal breaks. Don’t make a PVD. Learn to use the scissors in the diabetic traction detachments, particularly the more difficult ones, the so-called tabletop detachments. So what are the tricks? Avoid oil at all costs. Avoid retinal breaks at all costs. And use curved scissors and the advanced cutting technologies. Now we’ll go back to macular holes. How do I manage a 500-plus micron macular hole? Macular patch grafts, as I just described. So we’ll go through that technique again. We make a blister, a bleb, with a 38-gauge cannula with BSS, diathermize the vessels at the periphery of the bleb, cut it with scissors, leaving one little tag on the inner edge, do medium term… Excuse me. Put PFO in place with the MedOne dual-bore cannula, take the forceps, drag the graft flat against the retinal surface, position it properly, and leave the PFO in for a week. So that’s how I manage big holes. What about stem cells? First of all, embryonic stem cells, I think, are never gonna happen. I’m well aware that some people have published papers. I don’t approve. Not on moral grounds, but on the fact that these patients must be immune suppressed! We don’t have the capability to tissue-match these patients, and the availability of cells is just a huge challenge. So if we’re talking about stem cells, first of all, what layer are we talking about? To date, we’ve only talked about the RPE, and I recommend IPSCs. Well, what IPSCs? Well, they have to be 110, 120-day process. There’s over 25 tests that have to be done to make sure there’s no oncogenes or helper viruses or exons or other contaminants, and to make sure that they generate an EOG, that they generate a pump mechanism. All those tests have to be done, to make sure they’re — as Kapil Bharti and Juan Amaral at the National Eye Institute call — that they’re authentic RPE cells. So there’s some purely fraudulent work going on in Florida and other places about so-called stem cells. Mesenchymal stem cells derived from liposuction simply cannot repair retina or pigmented epithelium. So that’s a joke. But if it can’t repair the RPE, what about suspension of cells? This simply does not make a functional pigmented epithelium. It makes a lump of cells underneath the retina. What about iris RPE cells? They don’t support the visual cycle. What about the patient’s own peripheral RPE? Some believe that there are some epigenetic factors that mean it doesn’t properly support the macula. Patients that had ROP, that have an eccentric macula, as Bill Tasman reported years and years ago, lose vision later in life, because there’s something specific about the pigmented epithelium that’s under the macula, as opposed to peripheral retina — so, in short, not only will iris RPE cells not work; peripheral RPE cells are questionable. Cell suspensions are not useful. Embryonic stem cells, you have the problem of tissue — of rejection, and therefore it’s not an immune-privileged site. So we’re talking about IPSCs on a scaffold as being the only way to do it. And although there have been some recent reports, in my view, they were grandstanding, because they removed subretinal hemorrhage and then put the stem cells in place, and claimed that the stem cells made the patients see better. So we’ve got a ways to go. So now we’re asked about how to avoid making a macular hole when the VMT roof is very thin. I really addressed that earlier, when I talked about cystic change, but again, it’s proper visualization, flat contact, two, staining with Brilliant Blue, three, peeling toward the macula, and you can choose to leave a little cuff of ILM right on the VM roof, if you want to, although I have not done that. Then there’s a broad question. It’s often easy to decide when to operate on retina cases. How do you decide when not to operate? That’s a great question, and it’s a complicated question, but let’s address it as best I can in the time remaining. If the patient has a profound systemic illness, particularly something that’s life threatening, clearly that comes first, whether it’s a cardiac or it’s cancer. Now, patients that are stable, that are symptomatic, that are fully informed about their medical status, as long as you don’t put them at medical risk by operating, then if it makes them more comfortable in the last months or years of their life, it’s still okay to operate. But we’ve got to make certain that we don’t add to their medical complications by doing lengthy procedures, putting the patient to sleep, or improperly managing them medically at the time of surgery. Now, thinking about ocular reasons, there’s so many. The classic question is: How long do you wait in a vitreous hemorrhage? And they’ll say: Well, what’s the condition of the retina? What does the ultrasound show? What’s the vision in the other eye? What’s the patient’s occupation? What’s their health? How old are they? What is their mental attitude? Does surgery make them nervous, or does losing vision in one eye make them more nervous? So there are many patient factors, medical, psychological, and the condition of the other eye. For example, a very, very dense vitreous hemorrhage, when the other eye is 20/200 from macular ischemia, you might operate at 20/200. I mean — excuse me — in a very, very dense hemorrhage — within a week or so of that hemorrhage occurring, if you know that eye to be a 20/30 eye. If it’s far and away the better vision eye, so the patient can get back to work. But switch the situation. 20/30 in the other eye. Now you’ve got a vitreous hemorrhage in an eye that was known to be 20/200. You’re not gonna get them back to work by operating in that eye, so you may choose to observe for a while. So I think that’s important. Next question is: Patient with cataracts and VMTS who is undergoing cataract surgery. Will you prefer pneumatic vitreolysis or Jetrea? Neither. I prefer… If you have a proper view, fix the macula first. If there’s 3+ NS or posterior subcap, do cataract surgery, being careful to manage — to measure the axial length from the pigmented epithelium with low coherence interferometry, and then come back later and do the VMT. Next one: How can I induce PVD easily in a young patient? There is no easy way in a young patient. So what do I do? If you’re a young myope, and you get an inferior so-called round hole detachment, a non-PVD detachment, or an inferotemporal dialysis, I do a core vitrectomy, make no attempt to forcefully create a PVD, I put perfluorocarbon in the eye, fulfill, put them on topical difluprednate, Durezol, and leave the PFO in for two weeks. And when I go back in — and of course, I laser the breaks. When I go back in two weeks later, to take the PFO out, a PVD has been created by flotation from the PFO, and now I can remove the vitreous to prevent further traction to retina. So this is a fantastic way to do it. Now, PVD creation should not be done by moving laterally with your cutter. You should pull back from the optic nerve head with vacuum only applied. So people say: What about the flow rate of 27? 27-gauge is just fine to make a PVD, because you’re using the cutter as a forceps, not a flow rate machine. You’re simply capturing vitreous. If there’s a flow, you’re not doing it right. And you can’t go right over the nerve head. Think Weiss ring — there’s no vitreous there. So you go at the outer margin of the nerve head and pull straight back. If you move side to side, you put shear on the vitreous space. That said, in a young myope, it can be challenging, and therefore, I don’t do it and make more tears. I use the medium-term PFO method I described. What is the key to becoming a great vitreomacular surgeon? Visualization, hopefully with Ingenuity going forward, but certainly with a flat contact lens. Two, steady hands. Don’t drink coffee before surgery. Three, Brilliant Blue. Four, don’t use scrapers of any kind or picks. And it says: Do I need to do ILM peeling in this scenario? It doesn’t say what scenario. So I already listed all the scenarios. Vitreomacular schisis, vitreomacular traction, lamellar, and full thickness holes, epimacular membrane, macular folds and the hypotony maculopathy. I do it in all of them. Well, that’s my last question, and we have five minutes left. Unless the others have questions, we’ll finish at this point. And thank all the listeners. Hopefully we’ll do more of these going forward. Maybe we’ll do a whole hour answering questions. Without a specific topic. But I thank you all for listening.

Download Slides


May 27, 2018

Last Updated: October 31, 2022

Leave a Comment