Surgery: Live Surgical Demonstration: Vitreoretinal (Part II)

DR CHARLES: Hi. I’m Steve Charles, in Memphis, Tennessee. So what I thought I’d do is to give a lecture on techniques for diabetic traction retinal detachment surgery. I go to meeting after meeting, as many of you do, and I hear constantly this concept: There’s no need for scissors anymore, with modern cutters. The notion that you don’t need to know how to use scissors technique is simply wrong. And so the first point I’d like to emphasize is that if you’re going to do diabetic traction attachment surgery, avail yourself of the fastest cutting rates, the beveled cutters, the better fluidics, and it’ll help. And you can do most of it. Epiretinal membrane, most of the traction detachment. Think of: if I had to put in oil, I probably did a bad job in surgery. That’s rude to say, maybe. Some people might interpret it that way. But silicon oil does not support oxygen diffusion, and these retinas are oxygen deprived. And the results of diabetic traction detachment cases where oil had to be used are simply miserable, and you ought to think of every way to not make a retinal tear if you possibly can. If you try to view a tabletop traction detachment, try to remove densely adhered epiretinal membrane over the macula, you will make retinal breaks, and if they’re big enough, you’ll be pushed into using oil, and you’ll get a bad outcome. Lens management and cataract management are a nightmare in the context of oil. There’s no good way to do it. If you do a phaco, the capsule gets fibrous and you can’t YAG it with oil behind it, so you almost have to make them aphakic with the oil in, in these difficult cases. So think about: If I did oil, I didn’t do it right. I’m stunned when I hear people say the first step is to make a PVD. If you’ve got a diabetic traction detachment, never ever pull on the vitreous and make a PVD that way. You need to recognize that the posterior vitreous cortex is contiguous with epiretinal membrane. And so as you do the case, you will ultimately have a PVD. But don’t grab vitreous with the cutter and pull on it to make a PVD in a diabetic traction detachment. Because not only will you make tears at the outer edge of the arcades where the retina is thin and ischemic — you’ll make tears in the periphery and not even see it if there’s not any vitreous hemorrhage. Which is a terrible thing to do, because you’ll end up with postoperative detachment, and very bad outcomes, neovascular glaucoma, et cetera. And rows of laser in the periphery or an unnecessary buckle are not the solution. So let’s come back to… What should you do, then? If the vitreous — always bear in mind the continuity of the posterior vitreous cortex. It is attached to the outside margin of every epiretinal membrane. So the peripheral edge of it. And every epiretinal membrane area along the arcades — unless there’s a little defect in the posterior vitreous cortex over the macula, for the most part, there’s always vitreous between. So there’s no such thing as a vitreous band. Think about this continuity of the vitreous cortex. It has multiple layers. Put membrane peeling behind you. Occasionally the membrane will be easy enough that as you try to remove it with the cutter, it will free. Occasionally the PVD will present itself as you’re doing it, but don’t think that way. Think about: I’m gonna do this by sharp dissection, not by pulling. Pulling is dangerous. Just try to put it behind you, that notion. So the idea is that there’s some linear algorithm. First you do a core vitrectomy. There is no core in the diabetic traction. It’s a plate called the posterior vitreous cortex. Then you pull on the vitreous to make a PVD. Wrong. Don’t do that. Think about — go into the disc, the optic nerve head, where the epiretinal membrane extends along the arcades, and take the scissors and delaminate in an inside out direction, or the cutter if there are islands. Now, let’s talk a little bit about cutter delamination. To go back to my first statement, there is no question that the cutter with high cutting rates, beveled cutters, and better fluidics, that we’re able to do more and more epiretinal membrane — but how? There’s two discretely different methods. If the epiretinal membrane is a plate, if it’s thick, if it’s dense, if it’s inflexible, then you’ve got to hold the cutter right approximated to the edge. You’ve got to hold it just like so. And I call it conformal cutter delamination. What does that mean? That means that since the retina folds up and down, you’ve got to feed it in the edge. But you’ve got to have the little up angle. You’ve got to have the port uprotated like this, a little bit away from the retina, so you don’t suck membrane. Simply feed it in. And carefully approximating at the edge, rotating out. Now to your overall goal. The overall goal, at a minimum, is to get the macula attached. That’s where central vision resides, as we all know. Secondary goal, get the posterior pole attached. Epiretinal membrane isn’t cancer. Your goal isn’t to get it it all out. If there’s a little retina with epiretinal membrane, leave it alone. Managing the posterior vitreous cortex. Now to an often-used term, which is ridiculous: En bloc. En bloc means in one piece. En bloc is absolutely essential in cancer surgery so that you get the proper histology and cancer margins and you don’t spread cancer cells. It has nothing to do with diabetic traction detachment surgery. It’s a ridiculous term that never should have got in our field. The original paper on it said: Pull on the posterior vitreous cortex to lift up the epiretinal membrane. Gotta be the stupidest idea you ever heard. It’s a great opportunity to make retinal breaks in the periphery that you can’t find, and can’t treat, and result in retinal detachment and bad outcomes. So forget en bloc. Where does segmentation fit? I invented segmentation, which means cut the membrane into pieces, and then a couple of years later, I invented delamination. What’s the difference? Pieces. So the traction along the retina goes away. So the retina can… As the epiretinal membrane epicenter separates, the retina has enough release of traction to go back against the pigment epithelium. Delamination means put the scissors underneath the membrane and remove it, like so. So cut it up, cut it off. The recurrence rate is less with delamination and there’s less atrophic retinal breaks. So where does segmentation fit today? Segmentation fits as an access tool. So now what scissors should you use? I’ve been preaching to learn the skill set of scissors. Don’t try to do it all with the cutter. I’ve been preaching this for quite a while now. Where do the scissors fit? Use curved scissors. Don’t use so-called horizontal or vertical scissors. The blade width is greater in scissors. The back wall of the eye is curved. What about bimanual surgery? Unnecessary. I don’t understand why you have to have it. If you use scissors that properly shear, and you use them properly, you don’t need bimanual surgery. When you do bimanual surgery, you have to use a chandelier for illumination because you’re using both hands for access tools, and chandelier provides so much glare, particularly if there’s cataract or a multifocal IOL, and makes it very hard to see. So much better to use focal illumination with an endoilluminator so you can see what you’re doing. In this notion of access segmentation first, then inside out delamination, starting around the disc to free up the arcade so the macula will go on, instead of the term en bloc, which implies in one big piece — while it might look impressive in a video, it’s not what you want. Segment the epiretinal membrane into several pieces so you’re not causing remote traction when you cut. Now, how do you use the scissors? Don’t stick the scissors in the potential space between membrane and retina and open them. Don’t stick the scissors wide open and then close. You’ll tug on the retina and tear it. If you take the scissors and close them almost completely and just cut right at the tip, remember the scissors cut on a point. Not on a line. Sharp scissors. They’re square edges. They cut by shearing. And what you want to do with the scissors is to introduce them almost closed, right at the little attachment points, and sever them. What should you do about hemostasis? Well, hemostasis is obviously crucial. But if you use the diathermy extensively, as you’re doing this, you will create retinal necrosis, which results in retinal breaks. The instant you diathermize the retina, it gets weaker. Elevate the pressure to control any bleeding, and later, if the bleeding continues when you lower the pressure, use the laser for hemostasis. Laser hemostasis is far superior to using the bipolar diathermy for hemostasis in most instances. Why? Because the 532 nanometer green laser beam is absorbed in the red hemoglobin column, and energy is delivered just to the blood column and it coagulates. When you use diathermy, there’s a much broader spread of the radio frequency energy, hence necrosis, and patients later will develop large holes in posterior retina. Years later. And you’ll call them atrophic or ischemic to excuse yourself, but it’s actually created by the diathermy. Now, what about PRP? My recommendation is: if you’ve got elevated retina, which is the definition of diabetic traction detachment, don’t worry about reattaching it intraoperatively unless there’s a hole already present, or you make one. If there are no holes, don’t use gap. But never PRP that was detached pre-op. Why? We have anti-VEGF compounds you should have injected at a week pre-op, and you should inject the anti-VEGF compound at the end of the case anyhow. That’s a better way to fight neovascularization. If the PRP retina was detached pre-op, always some subretinal fluid there, and down the path of excessive retinopexy, excessive necrosis, excessive inflammation and fibrin syndrome. One program in the United States reported hundreds of cases of fibrin syndrome. In my mind, I trace that back to this notion of doing… In traction detachment, draining fluid, putting an air bubble in, and then doing a PRP. It results — and I’m repeating myself intentionally — in excessive retinopexy. So what are some of the crucial issues I brought up? Learn to use scissors. Always use curved scissors. Use them when the now updated methods of beveled cutters, 10,000, 20,000 cuts per minute, better fluidics, still make you concerned you’re gonna make a hole in the retina, particularly in the posterior pole, particularly in the macula. So keep that scissor expertise in your surgical armamentarium. Don’t pull on the posterior vitreous cortex and make a PVD and diabetic traction detachment. Please don’t do that. It’s not the right thing to do. If you use oil, be self-critical. Say “I made a mistake. I shouldn’t have done that.” There are some cases that are so bad that we have to, but think of every possible way to avoid breaks and avoid oil. As a sidebar issue, if you make a hole in an ROP case and put in oil, it’s over. You’ve blinded the eye. You’re not gonna fix it. Keeping these things in mind, that oxygen supply can’t happen in the presence of silicon. You can’t have diffusion from areas of perfused retina to areas of underperfused… That is such an important concept to keep in mind. So let’s talk a bit about preparation before surgery. It’s very, very important to work through the patient’s medications. Say what should you continue to take on the day of surgery. Obviously they should not eat or drink anything for eight hours or so. But blood sugar management is crucial. And so what often happens is patients elect to not take their insulin the day of surgery, and then they walk in with a blood sugar that’s too high to safely operate. If you try to acutely get the blood sugar down with insulin, you have concerns about potassium levels. So it’s better to do a better job and bring the patient in another day. Let’s talk a little bit about PVR. PVR, or proliferative vitreoretinopathy… Rhetorical question. Is PVR kind of like cancer? If you get it all out, it doesn’t come back? No. It’s not. PVR is a keloid. It’s excessive healing. What’s the difference between healing and scarring? It’s a functional definition. If you heal too much, it’s called scarring. Is the biology different? No. Do steroids get rid of — if you give them systemic steroids or subconjunctival steroids, is that gonna prevent PVR? It reduces inflammation, but most of the inflammation is because of excessive retinopexy, so that’s really not the answer. So at any rate, with PVR, think of it as a mechanical solution to make the retina reach. Sometimes that’s best accomplished by retinectomy. The late Robert Machemer recommended what’s called relaxing retinotomy, which means making a cut in the retina, in his case under infusion fluid, and then leave all the tissue anterior. At the same time he developed that, I developed retinectomy. What’s the difference? Retinectomy means remove all the tissue anterior to the cut. And so when I do retinectomy, I do it under air, because I’ve found that under saline, you can under or overestimate the amount of retinectomy required. So you start with removing the piece that represents the confluence of the anterior and posterior vitreous cortex in these PVR cases. Once you’ve resected that and brought it down to near the equatorial surface of the retina, then you address your attention to star folds. Starting in the posterior pole, the epimacular membrane is actually a star fold, if you think of it. Limited PVR. Take those with the forceps. Don’t get underneath the membrane with picks or things. Pinch-peel from the middle of the star fold and peel inside out. As you peel toward the periphery, often it will surprise you, and you’ll see a little star fold that will extend to a broad sheet of tissue that will peel right up to the equator, the posterior edge of the vitreous base, and stop. So peel as much as you can without damaging retina. Peel out to the equator. And then begin to drain through a retinal break. Then turn the air on. And as you do fluid-air exchange, continue your internal drainage. When the retina starts moving back, if it halts, if it stops moving back to the pigment epithelium, that means there’s either residual vitreous, in which case you should do vitrectomy under air, and failing that, if all the vitreous is properly removed and there’s no… What I call anterior loop traction, anterior to the equator, then something else is holding the retina up. Often if the retina was bullous, as two bulla come apart, you can see a star fold between them. Pull that off. So peel under air. As the retina continues to go back, then, if you say… Gee, it still won’t go back, but there’s no vitreous, there’s no membrane, what’s holding it up? Often at that point, particularly if there’s been turbid subretinal fluid removed, you’ll see a subretinal band and you can remove that band at that point. So think of it as the reattachment experiment, starting front to back. Did I get all the vitreous out? And again, you’re observing this under air. You’re observing this as you do internal drainage of subretinal fluid, which you initiated before fluid air exchange. So the retina begins to move back. Did I find vitreous? No? Oh, I did? Vitrectomy under air. Did I find membrane? I did? Peeling under air. Forget PFO. The retina goes back farther. It’s hung again. There’s a subretinal band tenting the retina. Let’s do punch retinotomy and remove the subretinal band, the retina goes back. Still won’t go back. It’s corrugated. Inferior vitreous base along the inferior equator. Still corrugated. So you make a little cut with the instrument and if subretinal air appears, that means the retina won’t reach. It doesn’t mean you did it wrong and you’ve got to start over. So continue to resect retina until you get it reattached. I call it incremental retinectomy. Once you’ve done retinectomy on the inferior half of the retina, often it will extend up past the 3 and 9:00 position. Some people then extend it 360. I don’t, but if it’s as much as 270, I go ahead and do the other quadrant to make it a full 360 retinectomy. Then you exchange for silicon oil. Often in these very challenging cases, I don’t use laser. Excessive retinopexy is one of the causes of PVR. So if you’re gonna have to do miles and miles of laser, think twice about it, and use silicon oil for what I call rhegmatogenous confinement, for the purpose of retinopexy avoidance, reminding you that PVR is a keloid. It’s not cancer. It’s not like you get it all out and it doesn’t come back. I wish it didn’t say P, in PVR, as in proliferative. Because there are very low cellular turnover rates. It’s a keloid. So your goal is to achieve mechanical reattachment with minimal tissue destruction, brief surgery, and the minimal causation of inflammation. That is crucial.