A glaucoma drainage device (GDD) is an implant also called an aqueous shunt or tube shunt. It consists of a silicone tube and a plate. The tube is inserted into the eye, and the plate is fixed onto the sclera near the equator of the eye. Aqueous humor is drained through the tube to the plate outside the eye, where it is absorbed by surrounding tissues through a fibrous capsule that forms around the plate 4 to 6 weeks after surgery 1).
It is positioned as a treatment option for glaucoma cases where conventional filtering surgery (trabeculectomy) fails to control intraocular pressure. The tube can be inserted into the anterior chamber or the vitreous cavity, depending on the presence of vitreous and the target intraocular pressure.
In Japan, two types of glaucoma drainage devices are available under insurance coverage (since 2012): Baerveldt® glaucoma implant and Ahmed® glaucoma valve. The Express® glaucoma filtration device without a plate is also approved 12).
For primary open-angle glaucoma (POAG), GDD is not recommended as a first-line surgery. It should be considered in reoperation cases. Evidence for normal-tension glaucoma (NTG) is insufficient, and further investigation is needed12).
QWhich patients are indicated for tube shunt surgery?
A
The main indications are cases where trabeculectomy with antimetabolites has failed, or cases with severe conjunctival scarring due to previous surgery (recommendation grade 1B)12). It is also indicated for conditions where trabeculectomy is unlikely to succeed, such as neovascular glaucoma, uveitic glaucoma, and ICE syndrome. For primary open-angle glaucoma, it is not recommended as a first surgery but should be considered in reoperation cases.
Since GDD is a treatment for glaucoma itself, understanding the clinical features relevant to the indication is important. In cases with the following disease types or background factors, the success rate of trabeculectomy is reduced, and GDD is considered.
Background factors for refractory glaucoma requiring GDD surgery include the following:
History of multiple intraocular surgeries: Conjunctival scarring progresses, making bleb formation and maintenance difficult
Young age: Fibroblast activity is high, leading to strong wound healing response and high failure rate of filtration surgery
Secondary glaucoma such as uveitis and neovascular glaucoma: Inflammation or neovascularization destroys the bleb
Conjunctival scarring: Due to chemical trauma, Stevens-Johnson syndrome, multiple conjunctival surgeries, etc.
In the PTVT Study (comparison in first surgery eyes), the failure rate in the tube group was higher than in the TVT Study targeting reoperation eyes, which is thought to be influenced by the younger age and higher proportion of Black individuals in the study group12).
GDDs are classified into two types based on the presence or absence of an intraocular pressure regulating valve.
Valved
Representative: Ahmed® Glaucoma Valve (AGV)
Mechanism: A pressure-regulating valve is built into the plate. The valve closes at an intraocular pressure of 8 mmHg or less, preventing excessive filtration immediately after surgery
Advantages: Intraocular pressure reduction is achieved immediately after surgery. Tube ligation or Sherwood slit creation is unnecessary. Risk of hypotony is low
Disadvantages: Long-term intraocular pressure reduction is slightly inferior to non-valved type (maintaining IOP <18 mmHg at 5 years: approximately 50%) 15)
Priming: Before surgery, irrigation fluid must be injected from the tube tip to confirm valve function.
Non-valved
Representative: Baerveldt® Glaucoma Implant
Mechanism: No valve mechanism. The tube is completely ligated with an absorbable suture (Vicryl®) during surgery, and several Sherwood slits (temporary leakage holes) are created to ensure early partial outflow.
Advantages: Excellent long-term intraocular pressure reduction (maintaining IOP <18 mmHg at 5 years: approximately 70%). Achievable IOP is lower 15)
Disadvantages: There is a period of high intraocular pressure for about 1 month after surgery. Risk of hypotony is higher than valved type (4.5% vs 0.4%) 12)
Postoperative management: For about 1 month until the absorbable suture dissolves, glaucoma eye drops are used concurrently to manage intraocular pressure.
The list of plate-type GDDs approved in Japan is shown below 12).
Device
Model
Plate Area
Insertion Site
Notes
Baerveldt®
BG101-350
350 mm²
Anterior chamber
Standard size
Baerveldt®
BG103-250
250 mm²
Anterior chamber
Pediatric / short axial length
Baerveldt®
BG102-350
350 mm²
Pars plana
Eyes with prior vitrectomy
Ahmed®
FP7
184 mm²
Anterior chamber
Built-in valve, most versatile
Ahmed®
FP8
96 mm²
Anterior chamber
Pediatric, short axial length
As a plate-less GDD, there is the Ex-PRESS® glaucoma filtration device (total length 2.6 mm, lumen 50 μm, stainless steel). It is inserted into the anterior chamber from under the scleral flap. Contraindications include angle-closure glaucoma, uveitis, and metal allergy. MRI up to 3 Tesla is considered safe12).
The choice between Baerveldt® and Ahmed® is based on the following criteria.
Cases aiming for lower intraocular pressure: Baerveldt® is indicated. It offers superior long-term intraocular pressure control, even considering the risk of postoperative complications.
Cases where immediate intraocular pressure reduction is desired or hypotony is risky: Ahmed® is indicated. This includes aphakic eyes, eyes with sutured IOL, eyes with history of expulsive hemorrhage, and uveitic glaucoma.
Long-term intraocular pressure control depends on the plate surface area. A larger plate area leads to a larger surrounding fibrous capsule, allowing more aqueous humor to be absorbed. The double-plate Molteno provided better IOP control than the single-plate, while in the comparison of 350 mm² and 500 mm² Baerveldt, the 350 mm² model was superior1).
QShould a valved or non-valved device be selected?
A
Even considering the risk of postoperative complications, Baerveldt® (non-valved type) is chosen for cases aiming for lower intraocular pressure. Ahmed® (valved type) is selected for cases where immediate postoperative IOP reduction is desired or where hypotony is dangerous (e.g., aphakia, uveitic glaucoma). Meta-analysis shows that the mean postoperative IOP with Baerveldt® (13.2 mmHg) is significantly lower than with Ahmed® (15.8 mmHg)15), but the incidence of hypotony is higher with Baerveldt® (4.5% vs 0.4%)12).
Medicine (Baltimore). 2023;102(42):e35449. Figure 2. PMCID: PMC10589554. License: CC BY 4.0.
Intraoperative photographs showing the key steps of Ahmed glaucoma valve implantation in 9 panels (A–I). These correspond to the individual steps such as tube insertion, plate fixation, and patch grafting discussed in section “5. Surgical Technique and Complication Management.”
Site selection: The superotemporal quadrant is the first choice. If placement is not possible due to existing surgical wounds, the nasal or inferior quadrants may be considered, but the inferior quadrant carries a higher infection risk and the nasal quadrant is prone to ocular motility disturbances, so these should be avoided as much as possible.
Conjunctival incision: A fornix-based incision is made, extending to expose two adjacent rectus muscles. The conjunctiva and subconjunctival tissue are dissected as far posteriorly as possible.
Plate fixation: Both ends of the plate are inserted under two adjacent rectus muscles, and the plate is sutured to the sclera with nylon sutures at a position 8–10 mm from the limbus.
Tube ligation (to prevent hypotony): The tube is completely ligated with an absorbable suture (Vicryl®) 2–4 mm anterior to the plate.
Sherwood slit creation (to prevent hypertension): Several temporary venting slits are made in the tube with a needle or blade to reduce immediate postoperative high IOP.
Tube trimming: The tube is trimmed with a bevel-up cut to a length that allows 2–3 mm insertion into the anterior chamber.
Tube insertion: A 23-gauge needle is used to create a puncture track at the corneoscleral limbus parallel to the iris plane, and the tube is inserted into the anterior chamber.
Tube fixation and covering: The tube is fixed to the sclera with nylon sutures and covered with a patch material such as preserved sclera or cornea (recommendation grade 1A)12).
Conjunctival suture: The conjunctiva is sutured and covered with absorbable sutures to finish.
The basic technique is the same as for Baerveldt®, but the following points differ.
Priming: Infuse irrigation fluid from the tube tip and confirm that the pressure-regulating valve opens.
Conjunctival incision range: Because the plate size is smaller, it can be performed with a narrower incision than Baerveldt®.
No ligation or venting hole needed: Since the pressure-regulating valve prevents early postoperative hypotony, tube ligation or Sherwood slit creation is not necessary.
Intravitreal tube insertion method (pars plana tube)
Steroid eye drops: Continue for about 6 months while tapering.
Baerveldt® postoperative hypertensive phase: For about one month until the absorbable suture ligating the tube dissolves, intraocular pressure remains elevated around 20 mmHg. Manage IOP with glaucoma eye drops or oral medications; IOP often decreases after 1–2 months. Another method is to pass a nylon suture through the tube and remove it early to adjust IOP during the hypertensive phase.
To prevent early postoperative hypotony with non-valved devices, the following measures are taken.
Intraluminal stent: Insert a 4-0 or 5-0 suture into the tube lumen and remove it under slit lamp after capsule formation.
External ligation: Ligate the tube with an absorbable suture (7-0 or 8-0 Vicryl®). A venting slit may be added to allow early partial outflow.
Two-stage surgery: In the first stage, only the plate is fixed; after 4–6 weeks, once capsule formation has occurred, the tube is inserted into the anterior chamber.
Tube exposure: A GDD-specific complication occurring in 4.3–14.3% of cases7). Tube exposure occurs through conjunctival erosion and carries a risk of endophthalmitis, requiring early repair8). Patch graft addition or scleral tunnel coverage is performed. To prevent tube exposure, patch materials such as preserved sclera or cornea should be used, or the tube should be covered with a partial-thickness scleral flap (recommendation level 1A)12).
Hypotony: More common with non-valved devices1). The incidence of hypotony is 4.5% for Baerveldt® and 0.4% for Ahmed®12). If the anterior chamber depth is maintained, conservative management is possible; however, if there is lens–cornea touch, anterior chamber reformation with viscoelastic injection is necessary.
Tube obstruction: Obstruction can occur due to fibrin, iris, blood, or vitreous. For anterior chamber tubes, Nd:YAG laser is used; for vitreous cavity tubes, vitrectomy is performed6). In cases with corneal opacity, intraocular endoscopy is useful for identifying the cause and treatment6).
Diplopia and ocular motility disorders: Occur in about 5% of cases postoperatively. Placement of the plate in the superonasal quadrant should be avoided. After 6 months of observation, management with prism glasses or surgery is considered.
Corneal endothelial damage: If the tube tip is close to the corneal endothelium, long-term corneal endothelial cell loss and bullous keratopathy may occur1). In elderly patients, combined cataract surgery may be recommended.
Tenon cyst (encapsulation): Fibrous capsule around the plate thickens, causing IOP elevation. The frequency is 40–80% for AGV and 20–30% for non-valved devices10). It often appears as a hypertensive phase 3 weeks to 3 months postoperatively.
Giant bleb: Rarely, a giant bleb forms around the plate 5). It is classified into an anterior type (foreign body sensation, cosmetic issues) and a posterior type (eye deviation, diplopia; T2-weighted MRI is useful for evaluation) 5). There have also been reports of uveitis after GDD removal 11).
QHow often do complications of glaucoma drainage devices occur?
A
In the 5-year data from the TVT Study, the incidence of serious complications was 34% in the tube group and 36% in the trabeculectomy group 13). Major complications include tube exposure (4.3–14.3%) 7), hypotony (Baerveldt® 4.5%, Ahmed® 0.4%) 12), and diplopia (about 5%). The types of complications differ between the two procedures: tube shunt surgery is associated with more corneal endothelial damage and tube exposure, while trabeculectomy is associated with more bleb leakage, hypotony maculopathy, and bleb infection 13).
In the Glaucoma Clinical Practice Guidelines (5th edition), CQ4 provides the following recommendations regarding the choice between tube shunt surgery and trabeculectomy12).
Recommendation: When selecting between the two procedures, it is recommended to consider the treated eye, patient background, and the surgeon’s proficiency with each technique.
Strength of recommendation: Weak recommendation against performing one over the other.
Strength of evidence: C (weak)
There was no significant difference in intraocular pressure control between the two procedures, and no significant difference in the incidence of serious complications that impair visual function. However, complications related to hypotony and postoperative infection were more common with trabeculectomy, while implant exposure and corneal endothelial damage were more common with tube shunt surgery 12).
Baerveldt 350 vs TLE+MMC (previously operated eyes)
Cumulative failure rate: tube 29.8% vs TLE 46.9% (p=0.02)13)
PTVT Study (3 years)
Baerveldt 350 vs TLE+MMC (first surgery eyes)
Cumulative failure rate: tube 33% vs TLE 28% (no significant difference)14)
ABC/AVB pooled (5 years)
Ahmed vs Baerveldt
Baerveldt 13.2 mmHg vs Ahmed 15.8 mmHg (p<0.001) 15)
TVT Study (Tube Versus Trabeculectomy Study) is a multicenter RCT of eyes with prior cataract surgery or trabeculectomy13). At 5 years, the tube group had a higher success rate than the trabeculectomy group. IOP reduction, medication use, serious complications, and vision loss were similar, but the number of additional glaucoma surgeries was significantly higher in the trabeculectomy group (p=0.025) 13). Postoperative QOL assessed with the NEI VFQ-25 showed no significant difference between groups 12).
PTVT Study (Primary TVT Study) is an RCT of primary surgery eyes 14). At 3 years, success rates were similar between groups, but the trabeculectomy group achieved lower IOP with fewer medications 14).
ABC/AVB Study is a multicenter RCT comparing Ahmed and Baerveldt 15). Over 5 years, the Baerveldt group was superior in IOP reduction and medication reduction compared to the Ahmed group, but serious complications were fewer in the Ahmed group 15). The incidence of hypotony after Baerveldt surgery (4.5%) was significantly higher than with Ahmed (0.4%) (p=0.002) 12).
In the Ahmed vs trabeculectomy comparative study, cumulative success rates (at 41–52 months) were 69.8% in the Ahmed group and 68.1% in the trabeculectomy group, with no significant difference. Tube exposure was more common in the Ahmed group, while bleb leakage and bleb infection tended to be more frequent in the trabeculectomy group 16).
Intraoperative use of mitomycin C (MMC) in GDD has been examined in several RCTs but did not show improved success rates 1). Prolonged hypotony and increased complications have been reported, and antifibrotic agents are not commonly used in GDD.
In pediatric glaucoma, angle surgery (goniotomy/trabeculotomy) is the first choice, but in secondary glaucoma its efficacy is limited, and GDD may be used as primary treatment 3).
Stallworth et al. conducted a systematic review and meta-analysis of 32 studies (1,221 eyes, 885 children) 3). The mean preoperative intraocular pressure was 31.8 ± 3.4 mmHg. The pooled mean intraocular pressure at 12 months postoperatively was 16.5 mmHg (95% CI: 15.5–17.6), with a success rate of 0.87 (95% CI: 0.83–0.91). At 24 months, the mean intraocular pressure was 17.6 mmHg, and the success rate was 0.77 (95% CI: 0.71–0.83). The success rate decreased to 0.54 at 48 months, 0.60 at 60 months, and 0.37 at 120 months. There was no significant difference in success rates between Ahmed and Baerveldt at 12 and 24 months. The most common complications were shallow anterior chamber (13.6%), hypotony (11.7%), and choroidal detachment (8.3%). Ninety percent of the studies used Ahmed, and data on Baerveldt in children are limited.
In children, the risk of tube-plate exposure is high. This is attributed to frequent eye rubbing and a strong immune response. In pediatric secondary glaucoma, especially after cataract surgery, surgical outcomes are poor, and GDD may eventually be required 12).
QWhich is better: glaucoma drainage device or trabeculectomy?
A
The optimal choice depends on the patient’s condition. For primary open-angle glaucoma as initial surgery, the PTVT Study showed that trabeculectomy achieved lower intraocular pressure14). On the other hand, in eyes with previous surgery or secondary glaucoma, the TVT Study favored tube shunt surgery 13). The glaucoma practice guidelines recommend selecting based on the treated eye, patient background, and surgeon’s expertise 12).
A tube covering method using an autologous scleral tunnel that does not require a patch graft has been developed.
Tanito et al. created a microincision scleral tunnel (MIST) using a 1 mm crescent knife and applied it for tube insertion into three sites: the anterior chamber, ciliary sulcus, and vitreous cavity 4). It is sutureless and can shorten surgical time, and no postoperative tube exposure was observed.
Miura et al. reported a scleral tunnel creation method using a 22-gauge needle, with no tube exposure observed during a follow-up of up to 21 months 7).
These techniques have the advantage of avoiding difficulties in obtaining patch graft materials and the risk of viral infection.
Kawashima et al. identified obstruction of the tube tip by fibrous tissue using an intraocular endoscope in cases of AGV failure where slit-lamp observation was difficult due to corneal opacity, and achieved intraocular pressure reduction by removing the tissue 6). Endoscopy was shown to be useful for diagnosing GDD failure in cases with corneal opacity.
Katsev et al. inserted an AGV in a case with combined outflow obstruction and reduced aqueous production, where the therapeutic range of eye drops was extremely narrow, and stabilized unstable intraocular pressure by creating a compliant outflow pathway 9). Intraocular pressure was maintained at 8–10 mmHg without medication for 15 months.
Intraocular pressure reduction by GDD is achieved through the following pathways.
Tube: A silicone tube that guides aqueous humor from inside the eye to the plate.
Plate: Fixed near the equator of the eye. It provides a space to store the aqueous humor guided through the tube.
Fibrous capsule: Connective tissue formed around the plate 4–6 weeks after surgery. Aqueous humor is absorbed from the outer surface of the capsule into the surrounding tissues (Tenon’s capsule and subconjunctival tissue).
Effect of plate area: A larger plate area results in a larger surrounding capsule, allowing more aqueous humor to be absorbed.
Mechanism of Postoperative Course after Baerveldt® Implantation
Immediately after surgery: Since the tube is ligated with an absorbable suture, there is no aqueous outflow to the plate. Only minimal leakage from the Sherwood slit occurs.
Approximately 1 month after surgery: The absorbable suture dissolves, and aqueous outflow to the plate begins. Intraocular pressure starts to decrease from this period.
2–3 months after surgery: The fibrous capsule around the plate matures, and the outflow resistance of aqueous humor stabilizes.
Long-term: As capsular fibrosis progresses, outflow resistance increases and intraocular pressure may rise (Tenon’s cyst).
The Ahmed® valve is a unidirectional valve that utilizes the Venturi effect; theoretically, the valve closes at an intraocular pressure of 8 mmHg or less. This prevents excessive filtration immediately after surgery and reduces the risk of hypotony-related complications (e.g., choroidal detachment, hypotony maculopathy). However, because the plate area is smaller than that of the Baerveldt®, the final achievable intraocular pressure is slightly inferior to that of the Baerveldt®.
European Glaucoma Society. European Glaucoma Society Terminology and Guidelines for Glaucoma, 6th Edition. Br J Ophthalmol. 2025;109(Suppl 1):1-230.
American Academy of Ophthalmology. Primary Open-Angle Glaucoma Preferred Practice Pattern. San Francisco: AAO; 2020.
Stallworth JY, O’Brien KS, Han Y, Oatts JT. Efficacy of Ahmed and Baerveldt glaucoma drainage device implantation in the pediatric population: A systematic review and meta-analysis. Surv Ophthalmol. 2023;68(4):578-590.
Tanito M, Ohtani H, Ida C, et al. Tube Insertion of Ahmed Glaucoma Valve Using a Micro-incision Scleral Tunnel Technique. Cureus. 2024;16(12):e75899.
Ugarte R, Sugihara K, Sano I, et al. Two different types of giant bleb formation following Ahmed Glaucoma valve implantation. Am J Ophthalmol Case Rep. 2024;33:102008.
Kawashima R, Baba K, Matsushita K, et al. Intraocular Endoscopy Resolved Tube Occlusion of an Ahmed Glaucoma Valve. Case Rep Ophthalmol. 2021;12:706-711.
Miura Y, Fukuda K, Yamashiro K. A Novel Scleral Tunnel Technique for the Prevention of Ahmed Glaucoma Valve Tube Exposure. Cureus. 2025;17(2):e79290.
Mansoori T. Management of Ahmed glaucoma valve tube exposure after scleral patch graft shrinkage. Indian J Ophthalmol. 2022;70(9):3432.
Katsev B, Campagna G, Liu JC. Ahmed glaucoma valve implantation in a case of volatile intraocular pressure and preoperative hypotony. Am J Ophthalmol Case Rep. 2025;38:102285.
Goel R, Gaonker T, Arya D, Golhait P. Outcome of combined single-loop fixation of intraocular lens with Ahmed glaucoma valve implantation. BMJ Case Rep. 2023;16:e254240.
Kolipaka PG, Krishnamurthy R, Bagga B. Unusual sequelae of staphyloma following Ahmed glaucoma valve explantation and its management. BMJ Case Rep. 2021;14:e239585.
Gedde SJ, Schiffman JC, Feuer WJ, et al; Tube versus Trabeculectomy Study Group. Treatment outcomes in the Tube Versus Trabeculectomy (TVT) study after five years of follow-up. Am J Ophthalmol. 2012;153(5):789-803.
Gedde SJ, Feuer WJ, Lim KS, et al; Primary Tube Versus Trabeculectomy Study Group. Treatment outcomes in the primary tube versus trabeculectomy study after 3 years of follow-up. Ophthalmology. 2020;127(3):333-345.
Christakis PG, Zhang D, Budenz DL, et al; ABC-AVB Study Groups. Five-year pooled data analysis of the Ahmed Baerveldt comparison study and the Ahmed versus Baerveldt study. Am J Ophthalmol. 2017;176:118-126.
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(3):464-470.
Copy the article text and paste it into your preferred AI assistant.
Article copied to clipboard
Open an AI assistant below and paste the copied text into the chat box.
