Surgery: Pupillary Membranectomy and Repeat Penetrating Keratoplasty

This video demonstrates a repeat penetrating keratoplasty with a pupillary membranectomy surgery. There was a lot of cortex left behind during the previous surgery and most of it was cleared. The suturing technique was demonstrated while suturing the donor cornea.

Surgery location: on-board the Orbis Flying Eye Hospital in Hanoi, Vietnam

Surgeon: Dr. Mark Mannis, University of California Davis


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

DR MANNIS: So we’re going to remove the previous graft and do a pupillary membranotomy. So we’re going to put in a 7.5-millimeter corneal punch and a 7.75 donor. Now, you’ll notice that the center mark is not in the center of the previous graft, but, rather, in the anatomic center of the cornea, because the previous graft was not placed centrally. Okay. The donor in this case is a 60-year-old, with a cell count of slightly over 3,000 cells per millimeter squared. It has been preserved in Optisol media, which will keep it alive usually for a period of about 10 days. So there’s our donor material. Okay. We’re going to open now. So this is the Hessburg-Barron trephine. It’s a suction device, which allows us very controlled entry into the anterior chamber. So once again, it’s very important to get this right. We have suction. One, two, three, four, five, six, seven, eight, nine, ten. Suction off. Each quarter turn of the trephine is 66 microns. This is the highest risk point in a keratoplasty, because it is at this point that sudden changes can take place — particularly with regard to the fluid dynamics in the retina. Very possible at this point to get a choroidal hemorrhage. This is one of those situations in which we don’t exactly know what we are going to find. So what’s happened here is that Descemet’s membrane is completely detached, and remains in the eye. This is one of the very important potential complications. It’s very easy to leave Descemet’s membrane in the eye. Which of course ensures that the subsequent graft will fail. Let’s have a SuperBlade. So now we’re going to take out Descemet’s membrane separately. So this is I guess what you would call a Descemetorhexis. The other problem with retained Descemet’s membrane is that there is invariably a remnant, which may occlude the angle. So this patient is at great risk for glaucoma.

>> So, Dr. Mannis, we have a question. So in a failed PKA, if you’d like to do an endothelial graft, some surgeons, they don’t strip the endothelium. They just put the DSAEK. What is the reason? Why it is different from the regular DSAEK?

DR MANNIS: Well, usually the Descemet’s membrane in a failed graft is very atrophic. And the posterior surface is suitably irregular enough that it’s actually a favorable circumstance for adherence of an endothelial graft, unlike a previously untouched cornea, in which the posterior surface is generally very smooth. As you know, we take steps to roughen the posterior surface. When we do a DSAEK in a Fuch’s patient or in a patient with pseudophakic bullous keratopathy, in a graft, that’s really not necessary. So some people do not remove Descemet’s membrane. There’s also a hazard in removing Descemet’s membrane in a graft. It’s very easy to disrupt the graft-host interface. So we prefer not to manipulate it, if possible. You’ll also notice that there is a dense membrane — we are not sure whether this patient is pseudophakic or not. But we’ll find that out in just a moment. I think she is. Very careful. Because we don’t know what’s there. Okay. Vannas scissor, please. Okay. It appears that the patient is phakic. So we’re going to have to do an aspiration of the lens and place one of the implants. I will need the Simcoe needle, please. This is one of those situations in which we were not aware of the lens status, because of the density of the membrane. But it appears that the patient is phakic. And that the lens is disrupted. So at this point, we will do a lensectomy. Or an extracapsular extraction. And a lens implant. Otherwise the patient will end up with a flocculent opacity. Very interesting. I think actually the patient is pseudophakic. Just that a great deal of cortex was left.

>> Dr. Mannis, so I have the patient’s UBM right here. And you can actually see an IOL.

DR MANNIS: Yeah, I think the patient is in fact pseudophakic, but the front of the lens was covered with cortex. As you can see, there’s a lot of cortex here. I’m not sure what the circumstance was, but we’ll try to very carefully remove as much cortex as we can, without jeopardizing the lens implant.

>> Mark, can I ask a question?


>> I notice you’re not using a Flieringa ring. And I know that you often do that back home. So I’m just wondering: What leads to your decision of using a ring or not?

DR MANNIS: We usually use — well, we use a ring in all children, in all aphakes, and in patients in which we plan to do a triple procedure. At which time, the lens is going to be removed, and we don’t have either the support of a lens or a pseudophake. In this situation, we assumed that the patient was pseudophakic, based on the UBM and the history, but when I opened the eye, there was so much cortex that I thought she was actually phakic. So we didn’t choose to put a ring on, because if the patient’s pseudophakic, they still presumably have zonular support for the anterior segment. I might point out also that I can see — I don’t know if you can see it on the TV screen — but there’s also a great deal of cortex or opacity behind the pseudophake. Rather than risk vitreous loss, I will leave that, and that will have to be approached with a YAG laser in the postoperative period.

>> Sir, there’s a question in the audience.


>> They were asking how do you calculate the IOL?

DR MANNIS: In answer to your question about power calculation, over time a corneal surgeon begins to get an idea of what his average curvature outcome is. And I know that most of the time, I have about 45 diopters of curvature. So knowing that I usually have about a 45-diopter outcome, what I do is I get the axial length, I put in the corneal curvature of 45, even though it’s theoretical, and based on that, I choose the intraocular lens. And admittedly, it is not precise. But it’s relatively effective, in terms of accuracy. I think that’s about as clean as we’re gonna be able to get it. It’s amazing. It’s almost as if the lens — as if there were no cortical clean-up at the time of the first graft, because there was a huge amount of lens cortex here. So we still have a considerable amount of cortex behind the lens, but I believe, looking at that, that that can be taken care of with a YAG laser. So at this point, we are going to stop the cortical clean-up, and go to suturing on the graft. I’m now covering this with viscoelastic. And we’ll be ready to start the placement of the corneal graft. So I’m now placing the cardinal sutures. The Polack forceps, which you see here, is used only for the first suture in a keratoplasty. And this is the first of the four sutures which are the cardinal sutures, which fixate the graft in position. So these, in a sense, are the four most crucial sutures in the procedure. I use a slipknot, which is very useful. As opposed to 3-1-1. This is a knot done in opposite directions. And it gives you extreme control over the tension. So there’s my… I establish my tension. Once I establish the tension, I turn the knot 90 degrees, and that locks it. So the second of the cardinal sutures is not the most difficult, but it’s the most important. Because it will now fixate the graft left to right. So the goal here is just to get the tissue to appose. You don’t want it to get any tissue compression. Because that will induce astigmatism.

>> So how deep the stitch, Dr. Mannis, you aim for?

DR MANNIS: Well, you know, different corneal surgeons have different philosophies about depth. I generally aim for a pre-Descemet’s suture, generally of the same length on either side of the wound. However, particularly in cases of keratoconus, I’m very comfortable with full-thickness corneal sutures.

>> So, Dr. Mannis, is it important to do the opposite suture? Because some, they don’t do it. Do you feel that it’s still needed, once you do the suture, to do the opposite quadrant?

DR MANNIS: Yeah, I think that, by doing opposite sutures, you have more control over the ultimate spherical outcome. Whereas if you go in a circle, you’re never sure if you’re pulling enough opposite. So now we have half of the sutures in. I’m going to put a little more viscoelastic into the anterior chamber.

July 25, 2017

Last Updated: October 31, 2022

Leave a Comment