Nonpenetrating glaucoma surgery: European experts compare notes

With pressures still low in long-term follow-up, practitioners discuss variations of this surgical treatment modality for glaucoma.

Ocular Surgery News Europe/Asia-Pacific Edition convened a round table of experts to discuss nonpenetrating glaucoma surgery at the International Glaucoma Symposium in Barcelona. An international panel of ophthalmologists discussed their experience with nonpenetrating deep sclerectomy, viscocanalostomy and the techniques’ many variations.

Philippe Sourdille, MD, is the director of the Clinique Sourdille in Nantes, France. He can be reached at Le Chaigne, 16120 Touzac, France; +(33) 545-21-25-51; fax: +(33) 252-83-87-19; e-mail: philippe.sourdille@wanadoo.fr. Dr. Sourdille is a paid consultant for Cornéal.

We will present the round table in two parts, beginning in this issue. In this installment, the participants discuss anatomy and manual technique, the use of implants, the need for postoperative monitoring and the use of sutures. The next installment will feature discussion of antimetabolites, indications for surgery, complications and long-term results.

The moderator of the round table session was Philippe Sourdille, MD, of Nantes, France. The participants were Christophe Baudouin, MD, PhD, Roberto Carassa, MD, Alfredo Mannelli, MD, Giorgio Marchini, MD, Paul Palmberg, MD, and Richard Parrish, MD.

Manual dissection

Philippe Sourdille, MD: The purpose of this round table is to discuss nonpenetrating glaucoma surgery in all its variations. We have brought together a panel of surgeons with a wide range of experience and different methods of performing essentially the same surgical procedure. For an objective opinion, we have also included a "control group" of two American surgeons who do not perform nonpenetrating glaucoma surgery or viscocanalostomy.

The difference between the results of viscocanalostomy and of deep sclerectomy will be interesting, because obviously this is the same operation, done by nearly the same people. Those who have published information on viscocanalostomy say they have no external filtration, or very little. Surgeons doing deep sclerectomy, like myself and other colleagues here, claim that we do need subconjunctival filtration.This is a major question for all of us.

First, we shall discuss the manual technique. Let us speak about the anatomical features of non penetrating surgery. Dr. Baudouin, you have published important work on the anatomy of normal and abnormal trabeculum. Do you think we have to remove the external trabeculum routinely in nonpenetrating surgery?

Christophe Baudouin, MD: I have performed nonfiltering surgery since 1995. I believe “deep sclerectomy” is a bad name because in order for this technique to work correctly, it must be done at the level of resistance to aqueous humor filtration or outflow. This resistance is in the inner wall of Schlemm’s canal and the juxtacanalicular trabeculum. We must remove this part of the trabeculum.

If not, it will be only a sinusotomy, with removal of sclera and the roof of Schlemm’s canal. This may work for a while because of mild filtration, but it will not work well in the end.

We have done confocal microscopy and found that the part we remove is more or less always the same size width and depth. It measures about 30 µm in depth.

Christophe Baudouin, MD, PhD, can be reached at Auinze Vingts hospital, 28 Rue de Charenton, Paris 75012 France; +(33) 1-49-09-55-08; fax: +(33) 1-49-09-59-11; e-mail: arepo@worldnet.fr. Dr. Baudouin has no direct financial interest in the products mentioned in this article, nor is he a paid consultant for any companies mentioned.

It is absolutely necessary to remove this part of the trabeculum. If not, it is not really a trabeculectomy, and in fact, it’s not even a deep sclerectomy.

Dr. Sourdille: Perhaps external trabeculectomy would be a better name, as proposed by Thom Zimmerman, MD, PhD, in 1979.

Alfredo Mannelli, MD: I am not a basic scientist, I am just an enthusiastic surgeon. All I know is what I see at the microscope in the operating room and at the slit lamp. When I started performing nonpenetrating glaucoma surgery I did not remove the internal wall of Schlemm’s canal, and my results were not as good as the results that I am achieving now.

Of course, the learning curve could also have played an important role in the difference of results, but there is no doubt that under the microscope I can see how the outflow of the aqueous humor increases after removing the internal wall of Schlemm’s canal. The maneuver is easy, and I think it’s a crucial part of this surgery.

Roberto Carassa, MD: It is mandatory to get brisk aqueous flow coming through the membrane. In my opinion, this is obtained in most of our cases while we are doing the cleavage of the membrane and advancing the membrane. Only in cases where the aqueous is not percolating sufficiently will I strip the inner wall of Schlemm’s canal.

In other words, this is not done routinely. It is done only when aqueous is not coming through sufficiently.

Looking at both human eyes and rhesus monkey eyes, we observed that the morphological changes that are seen in the area of cleavage are mainly related to microholes in the anterior portion of the trabecular meshwork and at the junction with Descemet’s membrane. This means you can get sufficient fluid from the anterior portion of the trabecular meshwork. You really don’t need to get morphologic alteration within the inner wall of Schlemm’s canal.

What is mandatory is to get aqueous through. If you are not getting aqueous through, you need to do something more, such as stripping the internal membrane.

Roberto Carassa, MD, can be reached at Ospedale San Raffaele, via Olgettina 60, 20132 Milan, Italy; +(39) 02-2643-3591; fax: +(39) 02-2641- 2912; private surgery: +(39)02-7733-1542; fax: +(39) 02- 7631-1438; e-mail: carassa@tin.it. Ocular Surgery News could not confirm whether Dr. Carassa has a direct financial interest in the products mentioned in this article, or if he is a paid consultant for any companies mentioned.

Giorgio Marchini, MD: I think it is mandatory to remove the external trabeculum. The percolation is quite abundant after this maneuver. Dr. Sourdille did an animal study that demonstrates that the IOP is reduced to a low level following this maneuver. Therefore, I think it is mandatory to remove the external trabeculum.

Paul Palmberg, MD: Is it now thought that removing the inner wall of Schlemm’s canal is necessary to get flow to the trabecular meshwork, or is stripping up to Descemet’s membrane and having fluid come through what is thought to be the primary path, or do you feel that if you get one or the other, it’s fine?

Dr. Sourdille: Dr. Palmberg has asked something very important. Where does the aqueous percolate? Is it only through the trabeculum or through the peripheral Descemet's membrane?

We have done some work with Sheng Lim from Moorfield's, measuring the IOP on cadaver eyes with permanent inflow of aqueous in the chamber and continuous monitoring of the IOP. We have seen that when the internal trabeculum remains intact after removing the external trabeculum, we had an ultimate IOP lowering and an average IOP of 3 mm Hg to 4 mm Hg. Since low pressure at day 1 is such an important prognostic factor, we must do as much as we can during surgery to have the lowest possible IOP.

Based on experimental, surgical and clinical results, we have to do our best, which includes removing the external trabeculum in all cases, if possible. We know that we have to dissect in front of Schwalbe's line.

The problem of whether aqueous comes through the peripheral Descemet's membrane or through the trabeculum, or both, is more an anatomical discussion, not a clinical one, because we know that to treat early failures we need to have a large corneal window to be able to do the goniopuncture. What we have learned is that we have to dissect well in front of Schwalbe's line to have a place to put the goniopuncture, and not to heal if we do the goniopuncture too posteriorly.

So on this point, Dr. Carassa says it's not always necessary to remove the inner wall of Schlemm's, but all the other participants say we have to do that. This is a consensus that we need to reach in nonpenetrating glaucoma surgery.

Implants

Dr. Sourdille: What about re-sorbable implants, whether of collagen, of cross-linked hyaluronic acid or other material?

Dr. Carassa: I am mostly involved in viscocanalostomy and I’m still using Healon (sodium hyaluronate, Pfizer). I have no direct experience with solid implants to increase the possibility of a patent ostium postoperatively.

It’s an interesting idea. Robert Stegmann, from the very beginning, used small stents in order to maintain patent ostia in high-risk cases, even though there is no evidence of a big difference in the long-term results between the techniques.

I’m still using the Healon GV, which is less expensive.

Alfredo Mannelli, MD, can be reached at IMO Barcelona, C/munner 10, Barcelona 08022 Spain; +(34) 93-25-31-500; fax: +(34) 93-41-71- 301; e-mail: alfredomannelli@ hotmail.com. Dr. Mannelli has no direct financial interest in the products mentioned in this article, nor is he a paid consultant for any companies entioned.

Dr. Sourdille: Do you use Healon GV for economical reasons or clinical reasons, or both? You do not feel the need for implants?

Dr. Carassa: There are two facts. Economics is one of the reasons, but not the main one.

We have to admit, unfortunately, that there is a lack of evidence regarding the mechanism of action in viscocanalostomy. According to the evidence from a study we did with Paul Kaufman, with viscocanalostomy we have interesting morphological changes at the level of Schlemm’s canal. We looked at animals sacrificed 3 months after the procedure. All over Schlemm’s canal there were unhealed openings with traces of viscoelastics, and with hydration of the sclera on the peripheral portion. This may be involved in the mechanism of action of viscocanalostomy.

I still think viscocanalostomy is a different procedure when compared with deep sclerectomy. I’m not so fond of the nonpenetrating portion of the procedure. I’m mostly interested in the role that Schlemm’s canal and the trabecular meshwork plays in lowering IOP in viscocanalostomy.

Based on the little evidence that we have, I still agree with the idea that the viscoelastic plays a role in the final control of the IOP. I’m not convinced until this is shown by a controlled study, that an implant will make a difference in the outcome of the procedure.

Dr. Marchini: I agree with Dr. Carassa. I think viscocanalostomy is totally different from deep sclerectomy. I have experience with the use of implants with deep sclerectomy. We have performed a prospective study comparing deep sclerectomy with a hyaluronic acid implant and without the implant to trabeculectomy. The evidence is clear that the implant is important to maintain the space created by deep sclerectomy.

The results at 1 year were much better with implant than without implant. We obtained a 38% reduction in IOP with implant, in comparison to a 20% reduction without implant. With trabeculectomy we obtained 50% reduction of IOP at 1 year.

The point is that the implant is essential to maintain the intrascleral space, but it may not be sufficient in maintaining IOP control.

But another observation: When I reoperate in some cases after 2 and 3 years, I’ve still found the hyaluronic implant inside the eye. So I think the implant is relevant for this point.

Dr. Sourdille: Results have been published at 7 or 8 years, which clearly indicate that the outflow is better with than without the AquaFlow (STAAR) collagen implant device. Dr. Mannelli, do you have experience with the AquaFlow? What do you think of implants in general? Are they necessary for the success of this operation?

Giorgio Marchini, MD, can be reached at Lungadige Panvinio 1, Verona I-37121 Itlay; +(39) 45-80-72-340; fax: +(39) 45-80-72-025; e-mail: giorgio.marchini@ univr.it. Dr. Marchini has no direct financial interest in the products mentioned in this article, nor is he a paid consultant for any companies mentioned.

Dr. Mannelli: I have used three different types of implants, including the AquaFlow, but I have the most experience with the Corneal SKGel cross-linked hyaluronic acid implant. In my opinion, the use of an implant is mandatory in this surgery to achieve low target IOPs. It’s easy to place and mechanically prevents the collapse of the intrascleral compression chamber. It also facilitates the filtration of aqueous humor from this chamber to the subconjunctival space.

I do agree with Dr. Marchini. Even though we were told that the cross-linked hyaluronic acid implant is a short-term reabsorbable implant, I have been able to find remains of the implant 1 year after surgery. It lasts much longer than we were told, and this is an intraoperative finding. In 4 years, I’ve re-operated in only two cases of deep sclerectomy and in both I’ve found remains of the implant.

Dr. Baudouin: I don’t think it’s absolutely necessary to have any kind of drain. I don’t use a drain, and our results are similar to what is published in the literature — except in the report by Andre Mermoud, MD, who demonstrated that during the second and third year there is a significant difference. But in my opinion, it’s very important to remove the inner wall of Schlemm’s canal and the external trabeculum to obtain a high level of outflow.

The problem is that the failure of glaucoma surgery is not at the level of the scleral flap in most cases. It is at the level of episcleral tissue and subconjunctival tissue. In this case, the presence of the drain is probably not of striking importance in that way.

Dr. Sourdille: As Dr. Baudouin says, we have long-term results from Andre Mermoud, MD in Lausanne with the implant. With this and Dr. Marchini’s, Dr. Mannelli’s and my personal experience with implants, we can conclude that it is better with than without.

Postop monitoring

Dr. Sourdille: How do you check your patients postoperatively, and what do you think are the main IOP-lowering mechanisms?

Dr. Marchini: I check postop using biomicroscopy, gonioscopy and especially ultrasound biomicroscopy (UBM).

The UBM reveals three possible pathways of aqueous humor. The first is external filtration, which is the same as trabeculectomy. The bleb is of the same type, a low reflective bleb, but quite smaller in dimension. We know that this type is associated with good success with intervention during trabeculectomy. And about 60% of our cases have external filtration with this type of bleb.

The second is a sort of uveoscleral filtration; that is the name I use for it. It’s a sort of line of hypoechoic space under the sclera and the suprachoroidal space, but it’s quite different from the flat uveal detachment that we leave after trabeculectomy. This is also present in 59% of the cases.

Paul Palmberg, MD, PhD, can be reached at Bascom Palmer Eye Institute, Box 016880, Miami, FL 33101 U.S.A.; +(1) 305-316-6386; fax: +(1) 305-326-6474; e-mail: ppalmberg@ med.miami.edu. Ocular Surgery News could not confirm whether Dr. Palmberg has a direct financial interest in the products mentioned in this article, or if he is a paid consultant for any companies mentioned.

They are not always associated. When we have these two forms associated, the result is better. The pressure is a little bit higher than 16 mm Hg if we have only external filtration, and under 14 mm Hg if we have associated uveoscleral filtration.

And the third is a sort of filtration activity of scleral tissue, in some areas around the intrascleral lake. We’ve seen a sort of hypo-reflectivity around the scleral lake. I have some clinical evidence that in some cases with good IOP control, they have signs of intrascleral filtration. I have only found these characteristics in two cases where it has compounded IOP.

When we have all these three types of signs in one case, the control is much better. So I think this type of intervention obtains three modalities of outflow. The two main modalities, I think, are external filtration and uveoscleral filtration.

Dr. Sourdille: You really think that doing what we call a “proper” deep sclerectomy, you need subconjunctival filtration? Dr. Mannelli, does that reflect your experience?

Dr. Mannelli: Totally. In fact, I start to get uncomfortable whenever I see a very flat bleb in the eye of my patients. In these cases, even if the IOP is still controlled but at the upper level of my target, I usually inject 5FU and perform Nd:YAG goniopuncture to maintain the subconjunctival filtration as long as possible. I do believe that, also in this surgery, subconjunctival filtration is mandatory for long-term success.

Dr. Baudouin: Yes. My opinion is that in most cases, this is a filtering surgery. However, we all know that a trabeculectomy can work with a flat bleb. I also feel that other mechanisms may be involved in IOP lowering. But I’m not very confident when I have a flat bleb.

Dr. Carassa: There’s a difference between viscocanalostomy and deep sclerectomy. In our randomized, prospective studies of viscocanalostomy, looking carefully at the mechanism of action with UBM, blebs were seen in one-third of the cases, no more than that. By stratifying and qualifying the types of blebs, we found a correlation between elevated IOP and wide blebs. That is related to some intraoperative problems that we incurred while closing and tightly suturing the superficial flap.

In my opinion, viscocanalostomy seems rather unaffected by the external filtration. Still, judging by the UBM, I agree totally with Dr. Marchini. There is a hypoechoic space that often appears below the lake.

The final point that is unresolved is how Schlemm’s canal plays a role in lowering the IOP. As I said before, this was one of the reasons we started the monkey study. Thus far, it seems that Schlemm’s canal is playing a definite role.

Richard K. Parrish II, MD, can be reached at Bascom Palmer Eye Institute, 1638 NW 10th Ave., Miami, FL 33136 U.S.A.; +(1) 305-326-6474; e-mail: rparrish@ med.miami.edu. Ocular Surgery News could not confirm whether Dr. Parrish has a direct financial interest in the products mentioned in this article, or if he is a paid consultant for any companies mentioned.

Dr. Parrish: Has anyone done aqueous flow studies in the various types of postulated mechanisms? Do you think if there’s a bleb is it any different, or just the resistance is different?

Dr. Sourdille: I am not aware of such a study. Interestingly, with UBM we have learned that there are many cases of uveal reabsorption in trabeculectomy. There are recent publications on that.

There is not much difference between penetrating and nonpenetrating after some months. But the problems are the same, the fibrosis and failure problems are the same.

From a clinical point of view, one of the important things we have learned is that we have to follow patients carefully, on day 1, day 2, day 8, day 15, 1 month, 2 months, 3 months, because that is the time during which problems can occur.

We have learned that we must do goniopuncture early, as soon as we have an IOP elevation. We have to monitor these patients carefully for their conjunctival status. We have to use antifibrotic drugs early.

In the beginning we said, okay, with outpatient nonpenetrating glaucoma surgery, 1 year postop everything is fine. No complications.

It is true that we have few complications, but we have fibrosis problems and wound healing problems, and this must be carefully monitored during the first 2 months. I think we all agree on that.

Dr. Carassa: Yes, I totally agree.

Dr. Marchini: It is the same as for trabeculectomy. I totally agree.

Dr. Baudouin: It is the same as a good trabeculectomy, but during the first few weeks follow-up there is a higher rate of complications in trabeculectomy.

Sutures

Dr. Palmberg: Early discussions of this technique mentioned closing the outer scleral flap very tightly. Since you are now looking for filtration, do you put more fluid in the anterior chamber and actually want to see some drainage from the edges of your scleral flap to know it will form a bleb? Is your way of managing scleral flaps different from what I had earlier heard?

Dr. Marchini: Not at all, but I usually have external filtration immediately after I perform the surgery. I use only two stitches, and not tight stitches. I have some immediate external filtration during deep sclerectomy.

Dr. Baudouin: I don’t want to have a tight scleral flap. For me, it’s not important because I want to have filtration; however, we have to be sure that we’re talking about the same thing. The philosophy for viscocanalostomy is totally different.

No stitches for me, because when I have no perforation, I have good filtration. This is my criteria.

Dr. Marchini: I need to stitch in order to maintain the space and to keep the implant in place. Only for those reasons.

Dr. Mannelli: I use two loose stitches in order to keep the implant in place.

Dr. Carassa: I use seven to eight stitches, using 10-0 nylon, very tight, avoiding perforation, avoiding external filtration. And then I use Healon GV within the scleral lake. Obviously, if you test the fluid, no fluid is coming up, just some of the viscoelastic. In one-third of the patients I get a bleb anyway.

Dr. Sourdille: There is no statistical analysis, but from one-third to two-thirds, or more, there certainly is a significant difference.

Dr. Palmberg: It sounds like what I’m hearing from you, that your practice now is not what I heard earlier about deep sclerectomy, where people were talking about closing tight. It wasn’t just the viscocanalostomy people, it was also deep sclerectomy. I’m wondering now if this is actually subliminally an important part of the operation — that you are, in fact, making sure you don’t close it too tightly, and maybe even look to see that there’s flow, and maybe do some moderation of what you’re doing if you don’t see it.

Dr. Sourdille: Whether we are ideologically right or not is not important. To be clinically right is more important, especially for patients. Again, this is what we learned from our failures, and from our success as well. I think we all agree, including Dr. Carassa, we have different approaches between viscocanalostomy and deep sclerectomy. But we agree on the fact that the postop monitoring of patients is much more critically important than we thought at the beginning because we have to watch all the possible IOP-lowering mechanisms, including subconjunctival filtration.

If it depends on uveal reabsorption only, there is little to be done. If it is Schlemm’s canal, there is little to be done. If it is subconjunctival, there is much more to be done. This is why I wanted to have this discussion, knowing that we would not agree on all points. But we are doing fine, at least in the discussion.

Dr. Carassa: It’s interesting to see that in the failure cases in viscocanalostomy, the procedure is managed similar to penetrating surgery. We have cases in which during the first 1 to 2 weeks we get an increase in pressure. The management of these cases is exactly like deep sclerectomy: suturelysis, needling the flap, elevating the flap with a needle and initiation of 5-FU injections, and eventually I do a goniopuncture. So it goes from viscocanalostomy to deep sclerectomy to a penetrating surgery, and in this way, you recover most of the cases in the early postop time.

Dr. Sourdille: Dr. Carassa, you said something important: that you monitor and treat the conjunctiva and suprascleral space before doing goniopuncture.

Dr. Carassa: Yes, but it’s more complicated. It depends on what we are seeing. I follow all patients with UBM, and I think it’s important to see what’s going on inside the lake.

If the lake is fully patent, and you have increasing pressure, and most of the time you can see a hyperechoic membrane shining on the UBM, the first step is goniopuncture. If you are facing flattening of the intrascleral space, you have to treat the intrascleral space and promote external filtration. The postop management is really individualized.

Dr. Sourdille: Again, you highlight the importance of careful postop monitoring and individual treatment.

Gonioscopy does not indicate the IOP-lowering mechanism. But it can indicate convex or concave trabeculum, especially related to the presence or absence of an intrascleral implant or remaining viscoelastic. Sometimes in pigment dispersion syndrome you will see pigment migrating through the anterior part of the trabeculum, indicating that there is filtration, which fits well with recent anatomical work on porosity of the peripheral Descemet's membrane.

Gonioscopy is helpful for two things: to treat early synechiae or iris occlusion in the bleb, especially after trauma, or to do goniopuncture. Goniopuncture, I think we all agree, has to be as anterior as possible, as far as possible from the iris and iris root, and must be wide enough to be successful. That means a real hole. And Dr. Carassa has noted that it must be combined sometimes with early antimetabolite treatment.

Editor’s note: The next installment of this round table discussion will appear in the August 2003 issue of Ocular Surgery News Europe/Asia-Pacific Edition.