The Tube Versus Trabeculectomy (TVT) Study was designed to answer the question of whether instillation of an aqueous drainage device (tube) or a trabeculectomy (trab) should be performed on patients who need glaucoma surgery and who have had previous cataract surgery or trabeculectomy. This is a multicenter, randomized, prospective trial that adheres to the tenets of good clinical trial design. However, as in other clinical trials and particularly in glaucoma, the interpretation of the results is confounded by the intricacies of design that limit their general application. There are 7 reasons why the findings of this study are unlikely to change clinical practice: 1) tubes are not better than trabs, especially when lower intraocular pressures (IOPs) are desired; 2) not all trabs are created equal: the relatively high rate of complications after trab in the TVT study is not a universal experience; 3) motility problems after tubes (especially the design used in this study) are frequent and serious; 4) follow-up is still relatively short, and tube erosion is a longer-term complication that is not yet fully evident; 5) the TVT study does not investigate eyes that have had no previous surgery; 6) the tube that was employed was the Baerveldt 350, and the results cannot be extrapolated to other tube designs that are commonly used; and lastly 7) established practice is often difficult to change.
The results of the first randomized clinical trial that compared trabs with tubes (in this study, the Ahmed glaucoma valve) as initial surgical management of glaucoma were reported in 2003. After 4 years of follow-up, the cumulative probability of success (defined as IOP <21 mm Hg and a ≥15% reduction from baseline IOP) was no different in the 2 groups. Several years later, a Cochrane Review concluded that “there is insufficient evidence to conclude that clinical outcomes of trabeculectomy differ substantially from those of aqueous shunts in similar patients with complicated glaucomas. There is also insufficient evidence to conclude that any specific aqueous shunt is superior to the others currently in widespread use.” The TVT Study was implemented to rigorously address this issue of “tubes versus trabs.”
The TVT Study has published its 3-year results. The primary success criteria included an IOP <21 mm Hg and ≥5 mm Hg and a ≥20% reduction from baseline. There was a statistically significant difference favoring the tube group ( P = .01). However, when success was defined as <14 mm Hg, there was no statistically significant difference between the groups. The authors’ conclusion: tubes had a higher success rate compared to trabs; though they produce similar reductions in IOP and number of glaucoma medications, the complication rate was higher after trabs. The published results indicate that when the target pressure is low, there is no difference between trabs and tubes, and there appears to be no difference between the 2 groups with respect to serious complications.
The rate of trab complications in the TVT Study is not consistent with our own experience or with previously published studies; that is, the complication rates seem high. For instance, early postoperative wound leaks occurred in 11%. Early wound leaks have been reported in other studies at 0% to 6%. The definition of shallow or flat anterior chamber may differ between studies and these rates may be difficult to compare, though the TVT rate of 10% seems high. It is several times the published rate of the UCLA experience in phakic patients and many times higher than that seen in pseudophakic patients. The TVT Study explicitly allowed releasable sutures to be used in the scleral flap, and presumably also allowed for laser suture lysis postoperatively. The advent of laser suture lysis and adjustable sutures has drastically reduced the incidence of significantly shallow anterior chambers postoperatively. With respect to later postoperative trab complications, bleb leak at 5% and endophthalmitis at 3% are also high. Late serious infections like blebitis and endophthalmitis have been reported in other studies at around 1%.
The high rates of shallow and flat anterior chambers, early wound leaks, blebitis, endophthalmitis, and bleb leaks in the TVT Study may be related to the high concentration and duration of mitomycin C use (0.4 mg/cc for 4 minutes). This large dose does not represent the mainstream parameters for the application of mitomycin C. For instance, I currently apply 0.25 mg/cc for 1 minute, having had greater complication rates at an earlier dosage of 0.3 mg/cc for 1 to 3 minutes. Not all trabeculectomies are created equal, and they should not uniformly be expected to be associated with the relatively high rates of complications reported by the TVT Study.
Motility problems with the Baerveldt 350 shunt (the plate has a surface area of 350 mm 2 ) are well known and have been published by the TVT group. The rates of motility problems at 10% in the tube group and 0% in the trab group during the first year are both statistically and clinically significant. Tube erosion is also a serious long-term problem. The TVT Study cited a 5% rate of tube erosion over the first 3 years, which will undoubtedly increase with longer follow-up. Tube erosion always requires surgery to repair. Corneal decompensation is an additional long-term problem with tubes. TVT cites a corneal decompensation rate of 10% in the tube group after 3 years. A recent study showed that tube placement significantly contributes to graft failure after penetrating keratoplasty. An additional and usually overlooked issue is the extra cost of tubes, in operating room time, the cost of the device, and the surgeon’s fee.
The TVT study does not address the important question of which procedure is best in previously unoperated eyes. Trabeculectomy is an effective and relatively safe procedure in such eyes. The majority of patients enrolled in TVT (174/212, or 82%) had previous conjunctival surgery in the form of either trabeculectomy, intracapsular or extracapsular cataract surgery, or scleral tunnel phacoemulsification. Tubes are already a well-established option for such patients. The learning curve of performing a successful tube and minimizing its complications is considerable and must be taught in residency and fellowship training programs. If this is done at the expense of teaching the performance of effective and safe trabeculectomy, the art and science of performing trabeculectomy may be permanently compromised.
Not all tubes are the same. The Baerveldt 350 may behave differently than other commonly used tubes, such as the Molteno and Ahmed. Long-term results have been reported with different Ahmed designs and have been compared with the Baerveldt in retrospective studies. The ABC Study (Ahmed Baerveldt Comparison) is currently underway to address this issue in a rigorous, prospective manner.
Finally, the sheer inertia of long-established clinical practice is often difficult to change. One example is the Glaucoma Laser Treatment Study, which showed that argon laser trabeculoplasty works as well as medicines in the initial treatment of glaucoma. However, this result did not cause most practitioners to abandon the long tradition of primary medical treatment. In the Advanced Glaucoma Intervention Study (AGIS), there was a significant race effect on the success of argon laser trabeculoplasty (ALT) vs trabeculectomy in patients with advanced glaucoma that was never translated into clinical practice. Factors other than the outcome of a single study also come into play, such as regional standards of care, consensus of glaucoma specialists, and general acceptance of the ophthalmologic community. Trabs will not be abandoned by most surgeons in favor of tubes because of the TVT Study results.
The TVT Study is an excellent example of a well-performed, prospective, multicenter, randomized trial that seeks to answer a very specific question, led by investigators with integrity who have carefully performed the study. The results are valid. But not all tubes are the same, and all trabs are certainly not the same! The results apply to a very narrow, specific set of conditions that cannot be extrapolated to all tubes, all trabs, all surgeons, and all glaucoma patients like the ones in the TVT Study. The findings of the TVT Study are not likely to change clinical practice, and probably should not.