Glaucoma is a progressive optic neuropathy in which elevated IOP remains the principal modifiable risk factor for visual-field deterioration. When medications, laser procedures, or prior surgery fail to provide adequate IOP control, incisional glaucoma surgery is indicated. Refractory glaucoma commonly includes eyes with prior failed filtration surgery, conjunctival scarring, previous cataract surgery, inflammatory glaucoma, neovascular glaucoma, aphakic or pseudophakic glaucoma, and other conditions associated with a high risk of trabeculectomy failure^[1].

Trabeculectomy with adjunctive MMC lowers IOP by creating a subconjunctival filtration bleb. It can achieve low IOP with limited dependence on topical medication but is vulnerable to bleb fibrosis, hypotony, wound leaks, infection, and late bleb-related complications. GDD surgery diverts aqueous humor through a tube to an endplate located beneath the conjunctiva. Tube shunts were historically used mainly in refractory glaucoma, although their use has expanded substantially^[2].

The most relevant randomized evidence for refractory glaucoma comes from the TVT Study, which compared a nonvalved tube shunt with trabeculectomy plus MMC in patients with prior cataract surgery and/or prior glaucoma surgery. Its findings provide a strong basis for procedure selection in surgically complex eyes^[1].

A narrative review was prepared using peer-reviewed randomized clinical trials and follow-up reports comparing tube-shunt surgery with trabeculectomy plus MMC. The primary outcomes extracted were: • Surgical failure • Mean postoperative IOP • Number of glaucoma medications • Need for repeat glaucoma surgery • Postoperative complications The TVT Study was emphasized because its population closely represents refractory glaucoma. The PTVT Study was included to distinguish outcomes in surgically naïve eyes from those expected in eyes with previous surgical intervention.

Surgical success and failure

The TVT Study demonstrated an advantage for GDD implantation in previously operated eyes. At 5 years, the cumulative probability of failure was 29.8% in the tube-shunt group and 46.9% in the trabeculectomy-with-MMC group. The adjusted hazard ratio favored tube surgery, indicating that trabeculectomy had more than twice the risk of failure in this population (hazard ratio 2.15; 95% confidence interval, 1.30–3.56; P = .002). Persistent hypotony contributed more often to failure after trabeculectomy, occurring in 13 trabeculectomy eyes compared with 3 tube-shunt eyes^[2].

These results support tube-shunt surgery as a preferred option in eyes with prior surgery, where conjunctival scarring and altered wound healing may compromise filtration-bleb survival. The lower failure rate is clinically meaningful because refractory glaucoma often requires durable IOP control over many years.

By contrast, the PTVT Study found no statistically significant difference in the cumulative 5-year failure rate between procedures in eyes without previous incisional surgery: 42% after tube surgery and 35% after trabeculectomy (P = .21; hazard ratio 1.31, 95% confidence interval 0.86–2.01). This distinction suggests that the relative benefit of GDD implantation is strongly influenced by surgical history and the anticipated risk of bleb failurE^[3].

Intraocular pressure control

Both procedures achieved clinically substantial IOP reduction. In the TVT Study, mean IOP at 3 years was 13.0 ± 4.9 mmHg in the tube group and 13.3 ± 6.8 mmHg in the trabeculectomy group, with no significant difference between groups (P = .78). The 5-year analysis similarly concluded that both operations provided comparable IOP reduction and maintained IOP in the low-teen range^[1].

In the PTVT Study, mean IOP at 5 years was 13.4 ± 3.5 mmHg after tube surgery and 13.0 ± 5.2 mmHg after trabeculectomy (P = .52). Therefore, when successful, both operations can achieve IOP values generally considered appropriate for moderate-to-advanced glaucoma^[3].

Medication burden

Medication requirements are an important practical difference between the procedures. In the PTVT Study, trabeculectomy achieved comparable IOP with significantly fewer medications at 5 years: 1.3 ± 1.4 medications after trabeculectomy versus 2.2 ± 1.3 after tube surgery (P < .001). At 1 year, the medication difference was also substantial, with means of 0.9 ± 1.4 medications in the trabeculectomy group and 2.1 ± 1.4 in the tube group^[3,4].

This finding may favor trabeculectomy in patients who have poor adherence, ocular-surface disease related to chronic topical therapy, financial barriers to medication access, or a strong preference to minimize medication dependence. However, this potential advantage must be weighed against the increased risk of failure in eyes with prior surgery.

Reoperation rates

Repeat surgery is particularly relevant in refractory glaucoma because each additional intervention may worsen conjunctival scarring and reduce future treatment options. In the TVT Study, the 5-year cumulative reoperation rate for glaucoma was 9% after tube-shunt surgery and 29% after trabeculectomy (P = .025). Most reoperations after failed trabeculectomy involved tube-shunt implantation^[2].

In the PTVT Study, the 5-year cumulative rate of additional glaucoma surgery was 18.0% in the tube group and 10.4% in the trabeculectomy group; this difference was not statistically significant (P = .15). These differing results reinforce that tube-shunt superiority is most apparent in patients with previous ocular surgery rather than as a universal finding for all glaucoma populations^[5].

Complications

The complication profile differs between procedures. In the PTVT Study, early postoperative complications occurred in 19% of tube-shunt patients and 34% of trabeculectomy patients (P = .013). Late complications occurred in 22% and 27%, respectively, without a statistically significant difference. Serious complications associated with vision loss and/or reoperation occurred in 2% of tube-shunt eyes and 8% of trabeculectomy eyes, although this difference did not reach statistical significance^[2].

Trabeculectomy-specific concerns include bleb leak, shallow anterior chamber, hypotony, choroidal effusion, blebitis, and endophthalmitis. Tube surgery can be associated with tube exposure, corneal endothelial cell loss, diplopia, motility disturbance, and device-related erosion. Thus, the choice is not determined by success rates alone; the risks most relevant to the individual eye must be considered.

The available randomized evidence indicates that GDD implantation is particularly advantageous in refractory glaucoma associated with prior cataract surgery, failed trabeculectomy, or prior glaucoma surgery. In the TVT Study, tube surgery reduced the absolute 5-year failure rate by 17.1 percentage points compared with trabeculectomy with MMC (29.8% vs 46.9%). It also reduced the absolute reoperation rate by 20 percentage points (9% vs 29%). These findings make tube surgery a compelling choice when durability is the principal objective^[2].

Trabeculectomy remains appropriate in selected patients. In surgically naïve eyes, it achieved similar long-term IOP control to tube surgery and required fewer topical medications in the PTVT Study. Its ability to obtain low IOP with less medication may be beneficial for patients with advanced glaucomatous damage who require a low target IOP, particularly when conjunctival tissue is healthy and intensive postoperative follow-up is feasible^[3].

Clinical decision-making should account for prior ocular procedures, conjunctival status, glaucoma subtype, inflammatory activity, lens status, target IOP, corneal endothelial reserve, likelihood of adherence to follow-up, and the patient’s tolerance for ongoing medication use. A patient with a failed trabeculectomy and extensive conjunctival scarring will usually be a stronger candidate for GDD implantation than for repeat trabeculectomy. Conversely, a patient without prior incisional surgery and with a need to minimize medication burden may reasonably be considered for trabeculectomy with MMC.

A limitation of this review is that GDD outcomes vary by implant type, particularly between valved and nonvalved devices, and outcomes may differ across glaucoma subtypes. In addition, the TVT and PTVT trials studied distinct patient populations; their results should not be interpreted as directly interchangeable.

In refractory glaucoma, particularly in eyes with previous cataract extraction, failed filtration surgery, or a high risk of conjunctival scarring, GDD implantation provides a more durable surgical outcome than trabeculectomy with MMC. The TVT Study showed lower 5-year failure and reoperation rates with tube-shunt surgery. Trabeculectomy remains a valuable procedure because it can provide similarly low IOP with fewer medications, especially in eyes without previous incisional surgery. The optimal procedure should be individualized:

• Favor GDD implantation in previously operated eyes, failed trabeculectomy, significant conjunctival scarring, or high risk of filtration failure.
• Consider trabeculectomy with MMC in surgically naïve eyes, when minimizing medication burden is important, and when close postoperative bleb management is feasible.