Treatment Outcomes in the Primary Tube versus Trabeculectomy Study after 1 Year of Follow-up

Edição: 05/2018

Purpose: To report 1-year treatment outcomes in the Primary Tube Versus Trabeculectomy (PTVT) Study.
Design: Multicenter, randomised clinical trial. 
Participants: Two hundred forty-two eyes of 242 patients with medically uncontrolled glaucoma and no previous incisional ocular surgery, including 125 in the tube group and 117 in the trabeculectomy group.
Methods: Patients were enrolled at 16 clinical centers and assigned randomly to treatment with a tube shunt (350-mm2 Baerveldt glaucoma implant) or trabeculectomy with mitomycin C (MMC; 0.4 mg/ml for 2 minutes).
Main Outcome Measures: Intraocular pressure (IOP), glaucoma medical therapy, visual acuity, visual fields, surgical complications, and failure (IOP of more than 21 mmHg or reduced by less than 20% from baseline, IOP of 5 mmHg or less, reoperation for glaucoma, or loss of light perception vision).
Results: The cumulative probability of failure during the first year of follow-up was 17.3% in the tube group and 7.9% in the trabeculectomy group (P = 0.01; hazard ratio, 2.59; 95% confidence interval, 1.20–5.60). Mean ± standard deviation IOP was 13.8±4.1 mmHg in the tube group and 12.4±4.4 mmHg in the trabeculectomy group at 1 year (P = 0.01), and the number of glaucoma medications was 2.1±1.4 in the tube group and 0.9±1.4 in the trabeculectomy group (P < 0.001). Postoperative complications developed in 36 patients (29%) in the tube group and 48 patients (41%) in the trabeculectomy group (P = 0.06). Serious complications requiring reoperation or producing a loss of 2 Snellen lines or more occurred in 1 patient (1%) in the tube group and 8 patients (7%) in the trabeculectomy group (P = 0.03).
Conclusions: Trabeculectomy with MMC had a higher surgical success rate than tube shunt implantation after 1 year in the PTVT Study. Lower IOP with use of fewer glaucoma medications was achieved after trabeculectomy with MMC compared with tube shunt surgery during the first year of follow-up. The frequency of serious complications producing vision loss or requiring reoperation was lower after tube shunt surgery relative to trabeculectomy with MMC.

The results of this study suggest that trabeculectomy with mitomycin C is a more effective procedure in surgery-naïve patients at 1 year, comparing to Baerveldt tube implantation. However, it was associated with a higher rate of serious complications. 

Trabeculectomy and tube shunt implantation (also known as glaucoma drainage device) remain the most commonly performed glaucoma surgeries worldwide. In the last years, several sources point to a decrease in the rate of trabeculectomies and an increase in the rate of tube shunt surgeries. The first landmark study that compared the two techniques was the tube versus trabeculectomy (TVT) study2. This study reported that tube shunt surgery provided similar reductions in mean intraocular pressure (IOP) when compared to trabeculectomy in patients that had had previous surgery (either cataract extraction with intraocular lens implantation or trabeculectomy). However, tubes were associated with higher success rates and fewer complications. These results have changed the clinical practice by widening the patient range for glaucoma drainage devices, in cases other than the ones in which trabeculectomy would likely fail, such as uveitic glaucoma, neovascular glaucoma or considerable conjunctival scarring due to prior surgery. However, after the 5-year results of the tube versus trabeculectomy (TVT) a question remained unanswered – If tubes have a similar IOP-lowering efficacy with higher success rates when compared to trabeculectomy in patients with previous surgery, what would the results be if they were compared in surgery-naïve patients?

The purpose of the Primary Tube Versus Trabeculectomy (PTVT) study was to compare the long-term safety and efficacy of Baerveldt tube shunt surgery (350-mm2) with mitomycin-augmented trabeculectomy in eyes that have not had previous incisional surgery. In short, what is the best first glaucoma incisional procedure?
For this, the authors conducted a multicenter randomized controlled trial (RCT), including a total of 242 eyes from 242 patients, from 16 different centers.The primary outcome was the cumulative failure rate. Failure was defined as IOP >21 mmHg or reduced <20% from baseline, IOP ≤5 mmHg, reoperation for glaucoma, or loss of light perception vision. Needling and anterior chamber reformation at the slit lamp were not considered reoperations for glaucoma. Secondary outcome measures included IOP, visual acuity (VA), use of glaucoma medical therapy, surgical complications, and visual fields.
This primary outcome seems to be a good option for a glaucoma surgical RCT since it carries relevant combined clinical information. On the other hand, having IOP or IOP reduction alone as a primary outcome (as was used in the TVT study) is now seen as less appropriate. This change is in accordance with what is seen across all medical fields, where the main outcomes should incorporate information relevant to the patient.

The inclusion criteria were:
• Age 18-85 years
• Glaucoma that is inadequately controlled on tolerated medical therapy with IOP greater than or equal to 18 mmHg and less than or equal to 40 mmHg
• No previous incisional ocular surgery

The exclusion criteria were:
• no light perception vision
• pregnant or nursing women
• narrow anterior chamber angle
• iris neovascularization or proliferative retinopathy
• iridocorneal endothelial syndrome
• epithelial or fibrous downgrowth
• chronic or recurrent uveitis
• steroid-induced glaucoma
• severe posterior blepharitis
• unwillingness to discontinue contact lens use after surgery
• previous cyclodestructive procedure
• conjunctival scarring from prior ocular trauma or cicatrizing disease precluding a superior trabeculectomy
• functionally significant cataract
• need for glaucoma surgery combined with other ocular procedures
• anticipated need for additional ocular surgery
• unwillingness or inability to give consent
• unwillingness to accept randomization
• inability to return for scheduled protocol visits

By choosing these criteria, the authors removed all closed-angle glaucomas and almost all secondary open-angle glaucomas, as well as potential causes for increased trabeculectomy failure rate, which are usually indication for primary tube shunt surgery. Also, patients that needed combined procedures (eg. phacoemulsification) were excluded since that would introduce a considerable bias in the results. Although the exclusion criteria state “narrow anterior chamber angle”, 8 patients with chronic angle-closure glaucoma (CACG) were randomized and treated.

Randomization was done for treatment allocation after stratification by age, race, presence of failed filtering surgery in the non-study eye and clinical center. This was done to ensure balance of the treatment groups with respect to the various combinations of those variables. Neither the patient nor the clinician were masked to the assigned treatment. No information is provided regarding other types of masking (measurement, data collection/analysis) so we have to assume that none existed, which might have led to bias, especially performance bias during follow-up measurements and management (which includes decision to reoperate).

Regarding the surgical techniques, all tube shunts were placed in the supero-temporal quadrant and all trabeculectomies were done superiorly. However, some heterogeneity existed in terms of type of conjunctival flaps (fornix or limbal based), techniques to restrict tube flow, tenonectomy and conjunctival closure. The authors state that “The method of temporary tube occlusion did not influence the rate of surgical success in the tube group. The cumulative probability of failure at 1 year was 18.2% in patients with intraluminal rip-cord sutures and 15.3% in patients with external polyglactin ligatures (p=0.86, log-rank test).”, but no information is given regarding the other surgical options (and results must be taken in with caution since this study was not powered for this analysis). It is also important to note that the mitomycin C dose used in this study (0.4 mg/ml for 2 minutes) is different from the one used in the TVT study (0.4 mg/ml for 4 minutes) and different from what is used by several practitioners in Portugal/Europe (0.2 mg/ml for 2 minutes). Additionally, the study protocol states that MMC can be placed “before or after scleral flap dissection, according to the surgeon’s usual practice” which can lead to bias.

Of the initial 242 randomized patients, 225 (93%) completed the 1-year follow-up period. Fourteen patients were randomized but did not receive any surgery (due to insurance issues, transfer of care, stroke, death and unknown reasons) and the remaining 3 did not complete the scheduled follow up. The authors state that an intent-to-treat analysis was done, which includes all patients who were randomized according to the randomized treatment, ignoring protocol deviations and withdrawals. However, since this is a surgical RCT they only analysed patients that received surgical treatment (removing the 14 that did not receive surgery after randomization). Two patients were randomized to the trabeculectomy group but received a tube, being analysed in their original group. Patients who underwent additional glaucoma surgery were censored from analysis after reoperation (since they were considered failures at that point), which is common in a survival analysis.

The average age was 61.4 years, two-thirds were male, half were African American, 90% had primary open-angle glaucoma (POAG), the mean IOP was 23.6 mmHg and the visual field mean deviation average was 14.6 dB. No significant differences in any of the demographic or ocular features were observed between treatment groups at enrollment. All eyes in the PTVT Study were phakic at enrolment. The 10% that did not have POAG, had CACG, pigmentary glaucoma, pseudoexfoliation glaucoma and “other”, with similar numbers in both groups.

Failure occurred in 17.3% of tube shunt eyes versus 7.9% of trabeculectomy eyes (p=0.01; log-rank test adjusted for stratum; hazard ratio, 2.59; 95% confidence interval, 1.20-5.60). Thus, tubes were approximately 2.5 times more likely to fail, when compared to trabeculectomy with mitomycin C. The most common failure criterion met was inadequate IOP reduction, closely followed by reoperation for glaucoma. The rate of reoperation due to glaucoma was 6.6% in the tube group and 3.5% in the trabeculectomy group (p=0.14, log-rank test adjusted for stratum). To evaluate for a possible selection bias (of doctors when deciding whether the patient needed further glaucoma surgery), the IOP levels were compared between treatment groups in patients who underwent glaucoma reoperation, as well as those who failed because of inadequate IOP reduction but did not undergo additional glaucoma surgery. In both cases, there was no significant difference.

Both surgical procedures produced a significant IOP reduction with a significant reduction in the use of medical therapy. However, a greater IOP reduction was achieved with fewer IOP-lowering medications in the trabeculectomy group. Mean IOP reduction in the trabeculectomy group was 46% (mean final IOP 12.4 mmHg) versus 37.5% (mean final IOP 13.8 mmHg) in the tube group (p=0.02), using an average of 0.9 versus 2.1 IOP-lowering medications, respectively (p<0.001). This also showed how tube shunts can lower IOP to the low teen level, with a mean final IOP below 14 mmHg. As for the success rate, trabeculectomy with mitomycin C had a higher (complete = with no topical IOP-lowering medication) success rate than tube surgery at 1 year (59% versus 14%, p < 0.001).

Recognizing that a target of 21 mmHg or less might not be clinically adequate for many patients, the research team also determined failure rates for lower target pressures. For IOP >17 mm Hg and reduced less than 20% from baseline, failure rates in the tube and trabeculectomy groups were still significantly different at 20.6% and 9.6%, respectively (p=0.012), while for IOP >14 mmHg, failure rates were not significantly different at 28.1% and 20%, respectively (p=0.15). The authors then conclude that “Because the differences in treatment outcomes were present using a range of IOP success criteria, the study results seem applicable to patients with the full spectrum of glaucoma from early to advanced damage”.

Postoperative interventions were performed with similar frequency in both treatment groups. The most common intervention in the tube shunt group was rip cord removal, and in the trabeculectomy group was laser suture lysis.

Safety in the PTVT study
Intraoperative complications were uncommon (8 patients) and did not significantly differ between treatment groups.
Early postoperative complications (<1 month after surgery) were more common in the trabeculectomy group, happening in 39 (33%) patients versus 25 (20%) patients in the tube group (p=0.03, chi-square test). Common complications included choroidal effusions, anterior chamber shallowing, and wound leaks. Wound leak and encapsulated bleb were more common in the trabeculectomy group and no complications were more common in the tube group.
Late postoperative complications were similar between groups, happening in 20 (16%) patients in the tube group and 18 (15%) patients in the trabeculectomy group.
Overall, the complication rate after surgery was higher in the trabeculectomy group (41% versus 29%), although not statistically significant (p=0.06). Additionally, the trabeculectomy group showed a higher rate of serious complications (surgical complications that produced a loss of 2 lines or more of visual acuity, required reoperation to manage the complication, or both) than the tube group (7% versus 1%, p=0.03). The same definition for serious complication was used in the TVT and Ahmed Baerveldt Comparison (ABC) studies.
It is important to understand that the low incidence rate of most complications precludes a proper statistical analysis and sometimes it is important to look at the clinical relevance instead. The authors correctly point out that hypotony maculopathy rates, although not significantly different between groups, were higher in the trabeculectomy group (5 patients versus 1 patient).

Reoperation for complications was more common in the trabeculectomy group. It was necessary in seven eyes with trabeculectomy (6%) versus only one eye with tube shunt (0.8%; p=0.02).
There was no significant difference in the rate of cataract progression between groups. “Cataracts were considered to have progressed if there was a loss of 2 Snellen lines or more that was attributed to cataract at the 6-month follow-up visit or thereafter, or if cataract surgery was performed”.
Also, there were no significant differences in visual acuity changes between groups.

According to the authors, there is only one other prospective randomized study comparing trabeculectomy with tube shunt surgery (with Ahmed glaucoma valve) in surgery-naïve eyes3. The authors found that the “surgical success, mean IOP, glaucoma medications, VA, and postoperative complications were comparable with both surgical procedures at the final follow-up period (41-52 months)”. However, this study included patients from Sri Lanka, with a younger average age (52 years old), both primary angle-closure glaucoma and POAG were allowed, mitomycin C was optional and the success criteria were similar but with a mandatory IOP reduction of at least 15%.

In conclusion, the results suggest that after 1 year trabeculectomy with mitomycin C is a more effective procedure, although with a higher complication rate, when compared to Baerveldt tube surgery. A greater IOP reduction and surgical success were achieved with fewer medications in patients randomized to trabeculectomy. However, early postoperative complications, serious complications and reoperation for complications were more common after trabeculectomy. In terms of success, these results support current practice, where in low risk unoperated eyes trabeculectomy is typically the first surgical option. Both the TVT and PTVT studies had similar inclusion and exclusion criteria, as well as failure definition. Despite some methodological differences, it seems that prior surgery is indeed very relevant when opting for a trabeculectomy. Results from the PTVT study are expected at 3 and 5 years of follow up. These follow up results will better show if the procedures have indeed a similar failure rate when the target is set for the low teens (IOP < 14 mmHg).

Note: a risk factor analysis was done (but not yet published; information collected from the American Glaucoma Society congress 2017) to understand which factors were related to failure. A lower baseline IOP (<21 mmHg) and being randomized to the tube group were correlated with a higher probability of failure. Most of these patients had a surgical failure because they did not meet the 20% reduction criterion, which is in line with current clinical knowledge that shows that tubes are usually less able to reduce low baseline IOP values.

Bibliography
1. Gedde SJ, Feuer WJ, Shi W, et al. Treatment Outcomes in the Primary Tube Versus Trabeculectomy Study after 1 Year of Follow-up. Ophthalmology. 2018;125(5):650-663.
2. Gedde SJ, Schiffman JC, Feuer WJ, et al. Treatment outcomes in the Tube Versus Trabeculectomy (TVT) study after five years of follow-up. American journal of ophthalmology. 2012;153(5):789-803 e782.
3. Wilson MR, Mendis U, Paliwal A, et al. Long-term follow-up of primary glaucoma surgery with ahmed glaucoma valve implant versus trabeculectomy. American journal of ophthalmology. 2003;136(3):464-470.​