Surgery: Trending Towards the Tube

While both tube shunts and trabeculectomies have advantages, choosing between them may come down to surgeon preference.

Joseph F. Panarelli, MD, Miami

Surgery is seldom our first choice when treating a glaucoma patient. Medical and/or laser treatment are initially employed in most cases, in an effort to lower intraocular pressure. However, if the IOP remains elevated and the patient is still progressing or is deemed likely to progress, then we’re faced with a decision about which surgery we should perform to keep the patient from losing more vision. That usually means choosing between a trabeculectomy and a tube shunt.

Is one a better choice than the other? Of course, to some extent the answer depends on the individual’s circumstances. Studies of practice patterns in recent years have found a clear trend towards surgeons performing more tube shunts and fewer overall trabeculectomies, especially in eyes that have undergone previous surgery. Selection of tube shunts over trabeculectomy has increased from 7 percent to 46 percent in eyes that have undergone previous trabeculectomy and from 8 percent to 45 percent in eyes with prior cataract extraction.1 A study by Pradeep Ramulu and colleagues, based on Medicare claims data, found that between 1995 and 2004 trabeculectomies in eyes without previous surgery or trauma dropped 51 percent, from 51,690 to 24,178 (although the number did increase 9 percent in eyes with scarring). During the same period, the number of tube shunts implanted increased 184 percent, from 2,728 to 7,744.2 

Surgeons are clearly becoming more comfortable with the tube shunt option. They’re no longer relegating tube shunts to the treatment of refractory glaucoma and patients at high risk of filtration failure.

The Evidence: Tube vs. Trab

In the first prospective, randomized trial of glaucoma drainage implants versus trabeculectomy, the Ahmed glaucoma valve and trabeculectomy for primary surgery were compared in 123 patients with average follow-up of 31 months.3 The mean IOPs and adjunctive medications were comparable in the two groups. No statistically significant differences between groups were found for visual acuity, visual field or short- or long-term complications. (Differences in low-frequency complications would probably have been undetected in this study.) The cumulative probabilities of success were 68.1 percent for trabeculectomy and 69.8 percent for the Ahmed valve.  

The Tube Versus Trabeculectomy study was a multicenter clinical trial that compared the 350-mm2 Baerveldt glaucoma implant to trabeculectomy with mitomycin-C in patients who had undergone previous cataract extraction with intraocular lens implantation and/or failed filtering surgery. The five-year results showed that patients who underwent tube shunt surgery had a higher success rate than the trabeculectomy group; cumulative probability of failure was 29.8 percent in the tube group and 46.9 percent in the trabeculectomy group.4 (Failure was defined as an IOP >21 mmHg or not reduced by 20 percent below baseline on two consecutive follow-up visits after three months; IOP <5 mmHg on two consecutive follow-up visits after three months; reoperation for glaucoma; or loss of light perception vision.)

In terms of IOP reduction, there was no statistically significant difference between the tube shunt group and trabeculectomy group. Both procedures produced a significant reduction in pressure that was sustained at the five-year follow-up, and there was a significant reduction in the use of supplemental medicines in both groups as well. Early postoperative complications were more frequently seen in the trabeculectomy group, though most were transient and self-limited. Late postoperative complications, reoperation for complications, vision loss and cataract extractions were not different between the two procedures. The one additional significant finding was that there was a higher rate of reoperation for glaucoma in the trabeculectomy group.4,5

Tube shunts have historically been used to treat refractory glaucomas—patients with advanced uveitic or neovascular glaucomas, or extensive conjunctival scarring, such as patients who’ve had multiple failed trabeculectomies. The inclusion criteria in the TVT study led to the implantation of tube shunts in patients at lower risk for surgical failure. What we can conclude from this study is that tube shunt surgery and trabeculectomy surgery will produce a similar reduction in IOP with a similar risk of serious complications in similar patient groups. 

The Evidence: Caveats

It’s important to critically review the results of the TVT study. There are three important points that should be addressed.

First, it’s important to consider the specific reasons that a greater success rate was reported for the tube group than the trabeculectomy group. In both groups, the majority of patients failed because of inadequate pressure control—but there were also other reasons for treatment failure. In the trabeculectomy group, a handful of people did not meet the success requirements because their pressure was too low. In this study, the protocol called for a mitomycin-C concentration of 0.4 mg/ml to be placed for four minutes. Many surgeons use MMC for a shorter amount of time or use a lower concentration of MMC, and may have lower rates of hypotony as a result. The authors also point out that hypotony may be an acceptable outcome of surgery if visual acuity is not affected. (In any case, when the three patients with hypotony and stable vision were reclassified as successes instead of failures, the study results did not significantly change.) 

Another major finding of the TVT study was that more patients in the trabeculectomy group had reoperations for glaucoma. This finding may have been skewed by the fact that surgeons have a greater comfort level doing a reoperation after a failed trabeculectomy (usually, putting in a tube shunt) than after a failed tube shunt. If a patient has an elevated IOP despite placement of one tube shunt, the surgeon may just increase medical therapy or observe a little longer before deciding to place a second tube shunt or perform a cyclodestructive procedure. On the other hand, it is important to note that in the TVT study there was no significant difference in mean IOP between treatment groups at the time of failure, in patients who had a reoperation for glaucoma or in patients who failed because of inadequate IOP reduction but did not have additional glaucoma surgery. These observations suggest that there was not a bias against reoperating for glaucoma in the tube group. 

Finally, the study is limited by factors common to many large, mul-ticenter, randomized clinical trials. Surgical outcomes can vary based upon the skill and level of experience of the clinicians involved in the study, and results cannot be generalized to different patient groups as this study enrolled patients who met very specific inclusion and exclusion criteria. Furthermore, low-frequency complications would have gone undetected (type II statistical error).

When I’d Choose a Tube

In my mind, there are three situations in which I would definitely choose a tube over a trabeculectomy. 

• High-risk glaucoma patients. The first situation is when I’m treating secondary glaucoma patients at high risk of trabeculectomy failure. These are patients who have a history of neovascular glaucoma, uveitic glaucoma, iridocorneal endothelial syndrome, epithelial downgrowth or aphakic glaucoma. I find these glaucomas very difficult to control, and trabeculectomies have a high risk of failure in these patient populations. In my experience, a tube shunt is often the best surgery for these patients.

• Patients likely to need future surgery. When trabeculectomies work, patients and surgeons are  happy, but when a bleb fails years later it can be disappointing. As we all know, additional surgeries performed on the eye create more inflammation and increase the risk of trabeculectomy failure. So, for example, I prefer to implant a tube shunt in any patient who has had a prior corneal transplant and needs IOP control. A trabeculectomy may initially work in these cases, but future surgeries for graft failure put the bleb at considerable risk. I have the same concern about patients who have serious diabetic retinopathy and may need retina surgery down the road. 

• When follow-up is questionable. When patient follow-up and  compliance is a concern, I’d much rather place a tube shunt to control the glaucoma. I don’t need to worry as much about these patients missing critical follow-up appointments in the early postoperative period—at least not as much as I would worry about a patient who’s had a trabeculectomy. A trabeculectomy patient needs a lot of postoperative care in the first few weeks, and missing a few appointments can be devastating to the formation of a good bleb. Suture lysis performed too late is often ineffective, and non-compliance with medical therapy (i.e., anti-inflammatory treatment) can also result in early bleb failure.

Does Choice of Shunt Matter?

Data from the Ahmed Versus Baerveldt Study and the Ahmed Baerveldt Comparison Study suggests that there are pluses and minuses to the two most popular shunts we have at our disposal. The Ahmed implant is great because it gives you the benefit of immediate pressure reduction, and the valve minimizes the risk of hypotony. However, you have to deal with the fact that these tubes can become encapsulated, in which case their ability to adequately control IOP decreases over time. 

One of the weaknesses of the Baerveldt shunt is that it is non-valved and hence needs to be occluded for the first few weeks after surgery. As a result, we don’t always get an ideal pressure reduction right away, despite utilizing different techniques to fenestrate the tube. When the tube opens, there’s also a higher risk of serious complications. Nevertheless, over the long run the Baerveldt seems to achieve slightly better pressure control (although the amount may not be clinically significant).

The bottom line is that it’s not yet clear whether one tube shunt is superior to the others. So for those opting to implant a tube shunt, the choice of device may come down to surgeon preference. Another factor is past experience; a bad experience with one device sometimes leads you to shy away from using it again in the future. A few cases of shallow anterior chambers with the Ahmed and you might switch to the Baerveldt. If you have some issues with hypotony after a Baerveldt opens, you might choose to switch to the Ahmed.
Ultimately, the results of comparative trials will have some impact on a surgeon’s choice of implant device, but training and clinical experience will also play major roles.

When I’d Choose a Trab
On the other hand, there are patients for whom I prefer a trabeculectomy to a tube shunt:

• Patients with severe disease. If a patient has more severe glaucoma, I’m likely to aim for a lower target IOP. In that situation, I tend to prefer a trabeculectomy simply because I can get a very low pressure with this surgery. I think most practitioners feel that tube shunts are a good way to lower IOP and are very predictable, but they don’t always get you a very low number. However, it was interesting to note that in the TVT study, the proportion of patients with an IOP ≤14 was similar between the two treatment groups. In fact, 63 percent of patients that underwent tube shunt surgery had an IOP less than 14.4 Since this seems to contradict traditional thought, it’s worth asking why. One possible explanation is that previous data has been skewed toward higher final pressures simply because in the past we’ve mostly implanted tube shunts in patients with refractory glaucomas who started out with very high pressures. 

• Phakic patients—especially those with shallow anterior chambers. I would prefer a trabeculectomy in this situation because there is the risk of endothelial damage from putting a tube in too close to the cornea. Since the patient is phakic, sulcus placement is not an option. I feel that trabeculectomy is generally the preferred surgical option in phakic patients, but it will be interesting to see the results of the ongoing Primary Tube Versus Trabeculectomy Study. Although there is a trend towards more tube shunts being placed, the number of total trabeculectomies being done still significantly outnumbers the total of tube shunts being performed.

Other Considerations

Usually, a surgeon’s three biggest fears with a tube shunt are tube erosion, endothelial damage and diplopia. The first two risks can be mitigated somewhat by the surgeon. For example, surgeons can decrease the risk of tube erosion by altering their technique a little bit. Tunneling the fistula tract a few millimeters prior to entering the anterior chamber and covering the tube adequately with a patch graft can help minimize exposure rates. 

Unfortunately, we don’t have great data regarding the effect tubes may have on the cornea, so most concerns about endothelial damage are based on anecdotal reports. One recent study looked at tube shunts after Ahmed implantation for refractory glaucoma, and it did find a reduction in corneal endothelial cells at two years of follow-up, with the greatest decrease in cell count in the quadrant of the tube’s insertion.6 However, another study found no difference in corneal decompensation rates when comparing pars plana tube placement to anterior chamber placement.7 

Furthermore, trabeculectomy and other glaucoma procedures can also decrease corneal endothelial cell counts. One comparative study found no differences in graft failure rates after trabeculectomy with mitomycin-C vs. glaucoma drainage device implantation.8,9 The PTVT study will compare endothelial cell counts in both trabeculectomy and tube shunt patients, and it will be interesting to look at the long-term results. 

The third concern, diplopia, is a real risk in patients with the Baerveldt implant. The TVT study found that the risk of developing diplopia was about 5 percent.4 Diplopia can be transient and often resolves within the first six months after surgery; if it persists, prisms can sometimes correct small deviations. However, there are times when patients ultimately need strabismus surgery for correction. 

Minimizing diplopia is a challenge because there’s not a lot we can do at this point. Diplopia is often related to the size of the bleb or the plate. To help decrease its incidence, about all we can do is make sure that our surgical technique does not disrupt or split the muscle fibers when implanting larger devices and ensure that the wings of the implant are completely beneath the proper muscle. Alternatively, we can choose an implant with a lower rate of strabismus postoperatively, such as the Ahmed glaucoma valve.10
Where Are We Headed?

I think it would be fair to say that when choosing between a trabeculectomy and a tube shunt for patients who have medically uncontrolled glaucoma, the best option is likely the one that you’re most comfortable with. The data from the TVT study suggests that the risk profile and IOP reduction are very similar between the two groups. Neither surgery showed a clear superiority. We’ve already seen a shift in practice patterns, in that we’re using tube shunts now for patients not only with refractory glaucoma but also those at lower risk for filtration failure. The question is, what role will tube shunts play for patients with mild or newly diagnosed glaucoma, or patients that have not had previous intraocular surgery? And, of course, new drainage devices are in the pipeline; how they will influence surgeons’ choices remains to be seen.

Minimally invasive glaucoma surgeries such as Trabectome, canaloplasty and the iStent are also gaining popularity. However, like any new procedure, these have a learning curve. Furthermore, these procedures may be better suited to treating mild-to-moderate glaucoma rather than more advanced disease. For patients with more severe glaucoma, I think a tube shunt or trabeculectomy remains the better choice; they produce a more significant reduction in IOP and help keep the patient from suffering additional visual field loss.

Many surgeons feel very strongly about whether a tube shunt or a trabeculectomy is more effective. You can make good arguments on either side, but what matters most is that both options work well.  REVIEW

Dr. Panarelli is an assistant professor of ophthalmology at Bascom Palmer Eye Institute in Miami.

1. Desai MA, Gedde SJ, Feuer WJ, Shi W, Chen PP, Parrish RK 2nd. Practice preferences for glaucoma surgery: A survey of the American Glaucoma Society in 2008. Ophthalmic surgery, lasers & imaging: The official journal of the International Society for Imaging in the Eye 2011;42:3:202-208.
2. Ramulu PY, Corcoran KJ, Corcoran SL, Robin AL. Utilization of various glaucoma surgeries and procedures in Medicare beneficiaries from 1995 to 2004. Ophthalmology 2007;114:12:2265-2270.
3. Wilson MR, Mendis U, Paliwal A, Haynatzka V.  Long-term follow-up of primary glaucoma surgery with Ahmed Glaucoma Valve implant versus trabeculectomy.  Am J Ophthalmol 2003 136: 464-470.
4. Gedde SJ, Schiffman JC, Feuer WJ, Herndon LW, Brandt JD, Budenz DL. Treatment outcomes in the Tube Versus Trabeculectomy (TVT) study after five years of follow-up. Am J Ophthalmol May 2012;153:5:789-803 e782.
5. Gedde SJ, Herndon LW, Brandt JD, Budenz DL, Feuer WJ, Schiffman JC. Postoperative complications in the Tube Versus Trabeculectomy (TVT) study during five years of follow-up. Am J Ophthalmol May 2012;153:5:804-814 e801.
6. Lee EK, Yun YJ, Lee JE, Yim JH, Kim CS. Changes in corneal endothelial cells after Ahmed glaucoma valve implantation: 2-year follow-up. Am J Ophthalmol Sep 2009;148:3:361-367.
7. Maris PJ, et al, J Glaucoma 2012 (epub ahead of print)]
8. Ayyala RS, Zurakowski D, Smith JA, Monshizadeh R, Netland PA, Richards DW, Layden WE. A clinical study of the Ahmed glaucoma valve implant in advanced glaucoma. Ophthalmol 1998; 105:10:1968-76.
9. Ayyala RS. Penetrating keratoplasty and glaucoma. Surv Ophthalmol 2000;45:2:91-105.
10. Hong, C. H., A. Arosemena, Zurakowski D, Ayyala RS. Glaucoma drainage devices: A systematic literature review and current controversies. Surv Ophthalmol 2005;50:1:48-60.