The Paul glaucoma implant: a narrative review and a perspective on design, surgical techniques, and current evidence

Introduction

Background

Currently, reduction in intraocular pressure (IOP) is the only proven method of preventing glaucoma progression (1). This can be achieved by medical treatment, laser, and/or incisional surgery. The vast majority of glaucoma surgery aims to augment the outflow of aqueous humor (AH). This can be accomplished via multiple approaches, including removal of the trabecular meshwork, opening up the suprachoroidal space to aqueous drainage, or most commonly, creating an alternative outflow route, draining AH from the anterior chamber (AC) to the subconjunctival space (external filtration). Trabeculectomy, the historical standard filtering procedure, achieves this by creating a fistula between the AC and the sub-Tenon’s apace, allowing AH to bypass normal outflow system, guarded by a partial thickness scleral flap that should provide the sufficient degree of flow resistance (2). However, trabeculectomy outcomes can be unpredictable and dependent on a number of factors including tissue response, surgical technique, and intensive postoperative care. Despite these limitations, trabeculectomy is very successful in reducing the IOP in eyes that have not had prior conjunctival surgery, though less successful in eyes that have either had prior surgery or in a range of secondary glaucomas. Other methods of external filtration have attempted to create a permanent drainage channel, using an adjunctive material or device. This was initially attempted almost a century ago using horse hair, followed by silk thread and other materials (3). The modern genre of aqueous shunts or glaucoma drainage devices (GDDs) date from 1973, when Molteno introduced the concept of draining AH away from the limbus to an end plate placed at the equator of the globe, where he hypothesized there would be less scarring than close to the limbus (4). The concept of equatorial drainage of AH is the basis of all plate based GDDs; shunting AH from the eye to a space-maintaining plate located in the equatorial subconjunctival space, via a tube, allowing free communication of the AC to a bleb, the area of which is determined by the size of the plate (5).

Since the 1970s, a number of GDDs have been developed. The most widely adopted of these; the Baerveldt glaucoma implant (BGI) (Advanced Medical Optics, Santa Ana, CA, USA) and the Ahmed glaucoma valve (AGV) (New World Medical, Inc., Rancho Cucamonga, CA, USA) were licensed in the early to mid-1990s. These have similar tube dimensions but differ in plate design and method of flow restriction. In addition to the equatorial capsule that takes 4–6 weeks to develop around the plate, the AGV is equipped with a flow restrictor that is intended to open when the IOP is 8 mmHg or above, thereby preventing hypotony and its potentially serious sequelae. In contrast, the BGI lacks a flow restrictor, so the surgeon must adjust the flow manually using an intraluminal rip-cord, one or more external tube ligatures, or a combination of the two. In a pooled analysis of the 5-year outcomes of two concomitant randomized surgical trials comparing these two implants, the Ahmed Baerveldt Comparison (ABC) study and the Ahmed Versus Baerveldt (AVB) study, the BGI demonstrated a lower failure rate and significantly better IOP control on fewer medications, albeit associated with a higher rate of hypotony (6). It was only recently that newer non-valved GDDs were developed, including the Paul glaucoma implant (PGI) (Advanced Ophthalmic Innovations, Singapore), released in 2017, and the Ahmed ClearPath (New World Medical, Rancho Cucamonga, CA, USA), approved by the U.S. Food and Drug Administration (USFDA) in 2019. The ClearPath has a plate design that is not unlike the BGI, and a similar tube portion to the BGI. From the limited literature available regarding the ACP, the safety and efficacy seem to be similar to the BGI, which is not unexpected (7).

Rationale and knowledge gap

Considering that PGI was essentially designed to address the safety/efficacy trade-off previously described with conventional GDDs, its usage in clinical practice is expected to either substitute the well-established GDDs or to expand to include other clinical situations. However, a reasonable body of evidence is required in order to justify larger scale adoption. Since its first-in-man clinical study (2019, including 6 eyes) (8), several reports on PGI outcomes have been published. While conventional GDDs (AGV and BGI) have been subject to a number of robust studies, including long-term randomized controlled trials, none of these were completed or published until well over 10 years after the devices were licensed for use. Similarly, high level evidence for the PGI, which has not been available for that period of time, is largely absent.

Objective

The purpose of this article is to provide a perspective on the PGI, highlighting its design, detailing the authors’ surgical technique (in standard settings and in various situations), and reviewing the literature that is currently available. This is expected to provide a useful guidance, especially in regard to surgical decision making in refractory glaucoma, where GDDs, including the PGI, is most needed. We present this article in accordance with the Narrative Review reporting checklist (available at https://aes.amegroups.com/article/view/10.21037/aes-2025-1-74/rc).

Methods

A literature search using a narrative review method was conducted using PubMed and Google Scholar databases. The end date of articles included in the review was August 2025. Search terms used included “Paul glaucoma implant”, “glaucoma drainage device”, and “aqueous shunt”.

For reporting PGI outcomes, we only considered English-published human studies that involved the PGI. Both comparative and non-comparative studies were included. Studies of follow-up duration less than 12 months, non-peer-reviewed articles, case reports, and conference abstracts were excluded. Both authors were involved in the selection process and consensus was obtained by discussion. Table 1 summarize the methodology employed.

The PGI

Design

Despite proven efficacy in IOP control and prevention of disease progression, widespread adoption of GDDs has been limited by a combination of unpredictability of early postoperative IOP control and a widespread belief that long-term efficacy is relatively poor. GDDs have been associated with a significant amount of morbidity including device exposure, corneal endothelial cell loss and diplopia, as listed in Table 2. In an attempt to reduce the risk of endothelial cell loss and exposure, the PGI was designed with a smaller tube diameter (127 µm internally and 467 µm externally) than prior GDDs (300 µm internally and approximately 640 µm for AGV and BGI), theoretically reducing the risk of corneal endothelial cell loss and tube exposure. Additionally, the smaller tube diameter appears to reduce the rate of post-operative hypotony, employing a smaller caliber ripcord (6/0 polypropylene) than the combination of ripcord and ligature(s) often used with the BGI (9). Conversely, a smaller tube might also be at a higher risk of internal obstruction with iris. Regarding the plate portion, the PGI also features a unique plate design, which, despite having a slightly smaller overall surface area than the BGI 101-350 (342.1 vs. 350 mm²), has a longer antero-posterior diameter at 16.11 mm (vs. 15 mm) and hence less area of the plate is tucked under the adjacent rectus muscles. In theory, this enlarges the effective filtration area that is not under the muscles (10). Different techniques in PGI implantation surgery will be discussed in the following section. Another unique PGI feature is the presence of a well at the junction of posterior end of the tube and the end plate. This allows direct visualization of AH flowing from the tube. The speed of filling the well is indicative of the rate of aqueous drainage through the tube, which can be subsequently titrated by either adjusting the intraluminal length of the ripcord or applying external ligatures, until a satisfactory filling is observed. This feature can potentially mitigate the IOP variability encountered during the early postoperative period.

The PGI is manufactured from medical-grade silicone and is available in one model/size. The PGI has obtained the CE mark in 2016, then it was released in 2017. The main characteristics of the PGI and other commonly available GDDs are illustrated in Table 3.

Surgical technique

Although there is a highly standardized basic PGI implantation technique at the authors’ institution, outlined below, there are variations in technique between and within surgeons. The following is an illustration of the authors’ most up-to-date technique, and how it can be adjusted for individual cases, succeeded by reviewing technique variations reported by others.

Routine PGI implantation

Sub-Tenon’s anesthesia is typically administered and it’s usually sufficient to achieve the desired levels of anesthesia and akinesia for PGI implantation, but general anesthesia is required occasionally. Unless occupied by other hardware (e.g., GDD or scleral buckle), the supero-temporal quadrant is the first quadrant of choice. Although some prefer supero-nasal, this is generally avoided with larger implants due to the risk of interference with the superior oblique tendon. In the case of a pre-existing GDD, especially with a large plate such as the BGI 101-350, the quadrant diagonally opposite (e.g., infero-nasal quadrant) is preferred. After placing a corneal traction suture(s), typically using 7/0 silk, the chosen quadrant is accessed through a wide limbal peritomy (around 3 clock hours). We routinely expose the equatorial sub-Tenon’s space to 2 sponges soaked usually in 0.5 mg/mL of mitomycin C (MMC) for 5 minutes, followed by irrigation with 20 mL of balanced saline solution. MMC is avoided in cases perceived to be at high risk of hypotony (e.g., severe longstanding uveitis), whereas higher concentrations of MMC (up to 1 mg/mL) are reserved for cases known or expected to scar aggressively.

The implant is then flushed to reduce static and ensure patency and a 6-0 polypropylene ripcord is inserted to restrict flow, initially for about 90% of the tube length. The wings of the plate are placed under the adjacent rectus muscles and sutured to sclera with two diagonally placed 9-0 polypropylene sutures, ideally at least 10 mm from the limbus. After trimming the tube bevel upwards to permit 2–3 mm of intraocular length, the tube is inserted into the AC or the ciliary sulcus via a 25G needle track, preferably at 12 o’clock, aiming to be as far from the cornea as possible, ideally just on the anterior iris surface, but not dipping into the iris stroma. If there is a pre-existing surgical iridectomy, this is a good position for tube insertion, minimizing potential tube related corneal injury and the likelihood of tube obstruction by iris tissue. It is important for the tube to be visible post-operatively. Therefore, when sulcus insertion is needed, we prefer to avoid blind insertion and prioritize tube visibility, even if that requires performing an overlying iridectomy.

After securing the tube portion to sclera, usually with two 9-0 polypropylene or nylon mattress sutures and ensuring that the AC is well-pressurized, aqueous drainage is examined at the well of the plate and the ripcord adjusted to permit minimal drainage. If the drainage appeared to be excessive, despite feeding the ripcord along the entire length of the tube, then 10-0 nylon ligatures are applied. In contrast, if insufficient drainage is observed, the ripcord is slowly withdrawn until an adequate level of drainage is observed. Occasionally, if no drainage can be elicited when the polypropylene has been almost completely withdrawn from the tube, or if the 6-0 polypropylene is too tight in the tube to thread in the first place, the 6-0 is replaced with a 7-0. For that reason, it is important to have 7-0 polypropylene available in the operating theatre for the occasional case in which it will be necessary. A donor patch graft is used to cover the anterior portion of the tube. We usually use fascia lata and occasionally donor sclera or cornea for tube patching. We have largely abandoned the use of pericardium because of its propensity to absorb and disappear quite rapidly. The donor patch graft and conjunctiva are fixed in place using fibrin tissue glue. The conjunctiva is also usually secured with 10-0 nylon sutures at the peritomy edges. The free end of the ripcord can either be left under Tenon’s to be accessible for partial adjustment and/or complete removal post-operatively, or, more conveniently, looped inside a limbal corneal groove, which allows easier access in clinic (11).

At the conclusion of surgery, betamethasone and cefuroxime are routinely injected subconjunctivally. Depending on the perceived risk of scarring, adjunctive medications e.g., intracameral bevacizumab (1.25 µg) and/or orbital floor/sub-Tenon’s triamcinolone are administered in selected cases. A routine PGI implantation surgical video can be accessed via https://www.youtube.com/watch?v=lkrso1E89UM&t=110s.

Post-operatively, topical corticosteroids (every 2 hours by day, gradually tapered over 3 months, and antibiotics (4 times daily for 2–4 weeks) are routinely prescribed.

In general, if the polypropylene ripcord has been adjusted intraoperatively to produce a roughly physiological rate of aqueous drainage, then early postoperative flow adjustments are rarely necessary. If this has not been achieved and the IOP is higher than ideal in the early postoperative period, then aqueous suppressants are usually added. IOP elevation resistant to aqueous suppressants is less common, but occasionally the ripcord can be partially withdrawn, assuming that the donor patch graft is not covering the entire length of the tube, and the ripcord length in the tube can be visualized. In general, the authors do not remove the ripcord in the first 3 postoperative months, especially if adjunctive MMC has been used during the implantation procedure.

If the opposite happens, i.e., the IOP is dangerously low in the early postoperative period, the authors’ preference is to inject small amounts of heavy cohesive viscoelastic at the slit lamp, e.g., 50–60 µL of Healon 5 Pro, due to its longevity. The common practice of using larger amounts of lighter cohesive viscoelastic devices such as Provisc or Healon Pro, results in more dramatic IOP elevation but shorter of duration, which we find less helpful.

If PGI implantation is combined with revising a prior glaucoma surgery (e.g., exploring an adjacent leaky trabeculectomy bleb, removing a pre-existing GDD, etc.) or performing concurrent surgery (e.g., cataract extraction or pars-plana vitrectomy), we typically insert the PGI first, adjust the aqueous drainage etc., then perform the other concurrent surgical procedure before conjunctival closure.

Technique variations

A group led by Professor Paul Chew, who designed the PGI and after whom the implant is named, uses a modified flow modulating method or “stability system”, in which a pericardial patch graft is inserted between the subconjunctival space and PGI plate with cross-linked viscoelastic (Healaflow) injected beneath the patch graft to fill the plate’s reservoir. This is said to alleviate the need for intraluminal stenting or ligation (12). However, early reports of this technique seem to indicate a higher rate of hypotony (n=17, 35.4%, at 3 years), albeit mostly transient and self-limiting. However, they also reported that 4 cases (8.3%) required AC reformation, and 1 (2.1%) developed persistent hypotony following late (1.5 years) blunt trauma to the eye. There were 4 additional cases (8.3%) of AC shallowing that was not associated with hypotony, half of whom required AC reformation.

Special situations

Occasionally, extra tube length is required if, for example, a tube retracts, requires repositioning, or is inadvertently cut during implantation or revision. There is no dedicated extender for the PGI. The Ahmed tube extender (New World Medical, Inc., Rancho Cucamonga, CA, USA), which was designed for the AGV but has also been used with other similar diameter tubes, is too large for a PGI (13). While a tight fit, the PGI tube can be connected to a tube larger diameter such as the BGI, though this can be technically challenging (see: https://www.youtube.com/watch?v=Fz3Pl03oLpo&rco=1).

Conversely, Berman and Au described the successful use of the PGI tube portion as a tube-in-tube extender for short BGI/AGV in 3 cases (14). It’s worth mentioning that the deliberate use of tube-in-tube method, not necessarily for short tube situations, has been also described, where a smaller tube that is typically of a non-plate-based implant (e.g., Xen gel stent or Preserflo microshunt) is inserted into the eye and connected to the larger BGI tube. Using this technique, the benefits of the great IOP reduction of an equatorially placed plate, the small intraocular tube, and the relative outflow restriction made by graded flow through 2 tubes of different diameters without ripcords or ligatures, can be obtained (15,16).

A pars plana clip was developed as a connecting segment for the AGV tube to facilitate insertion through the pars plana (17). A similar modified BGI (102-350), with the Hoffman elbow, has been employed to achieve that purpose during BGI insertion (18). In the authors practice, we have avoided using these as in our early experience, the small plastic plate close to the limbus had an extremely high exposure rate. In general, there is little advantage to pars plana modified GDD implants as the regular models can easily be inserted in the pars plana without difficulty (19), an impression that Zhou et al. shares (20). We believe that the key factor in achieving success with pars plana GDD implantation is not the angle of tube entry to the vitreous cavity, but rather performing an adequate vitrectomy, including vitreous base shaving by a vitreo-retinal surgeon, to minimize the possibility of tube obstruction with vitreous.

Outcomes

Efficacy

Regarding primary open-angle glaucoma (POAG), the most common glaucoma variant, the longest PGI follow-up results to date (3 years) have been reported by Tan et al. in 19 POAG cases (amongst 48 various glaucoma cases =39.6%) (12). For their entire cohort, they reported surgical success rate of 85.4% at 3 years, which was complete (without the need for glaucoma medications) in 75%. Most of the surgical failures were due to insufficient IOP reduction and occurred within the first post-operative year. The failure rate was higher in eyes with previous intraocular surgery, but it was similar comparing eyes that underwent PGI only to eyes in which a concurrent surgery, including cataract surgery (60.4% of study cases) was performed. At 3 years, the mean IOP was 14.9±4.1 mmHg, using 0.17±0.48 medications, achieving significant IOP and medication reduction at 5.69 mmHg (27.6%) and 2.96 medications (94.7%), respectively, from pre-operative values.

Additionally, the outcome of PGI in uveitic glaucoma, reported by Richardson et al., in 50 eyes for a mean follow-up of 35.8±9.8 months, was generally slightly better, with a success rate of 92% (complete in 48%) and a mean IOP of 12.2±4.4 mmHg, using 1.1±1.3 medications at the last visit, resulting in a significant percentage IOP reduction of 60.1% (21). This is comparable to other previously reported results from uveitic glaucoma that received 103-250 BGI implantation at an earlier time at the same center (22).

Furthermore, the largest PGI multi-center clinical trial was conducted by Koh et al. which included 74 eyes from 6 centers. They reported PGI success rate of 93.2% (complete success in 68.9%) at one year, along with a mean IOP of 13.2±3.3 mmHg on 0.3±0.6 medications, achieving 42.9% IOP reduction, compared to medicated pre-operative IOP (10).

Of note, these 3 PGI studies primarily used similar IOP cut-off values to define surgical success/failure. IOP values >21 mmHg or ≤20% reduction from baseline on 2 consecutive visits were required to consider insufficiently reduced IOP as a surgical failure. Additionally, success/failure rates at different IOP based criteria were also reported. A summary of the results of PGI studies that reported outcomes at least at 12 months post-operatively is illustrated in Table 4.

Safety

Similar to the BGI, the most commonly reported complications following PGI implantation were hypotony related. We mentioned earlier that Tan et al., using their “stability system” method, instead of a ripcord, experienced hypotony in 35.4% of cases. Moreover, 14.6% of the cases experienced what were considered “serious complications”, either because of the need for re-operation (n=4) (tube obstruction requiring intervention, 2; tube exposure, 1; corneal decompensation, 1), or being associated with a 2 or more line decrease in visual acuity (n=3), (choroidal detachment, 2; vitreous hemorrhage, 1) (12).

In the uveitic glaucoma cohort of Richardson et al., hypotony was reported in 5 (10%) cases, 4 of which occurred within the first 3 months and one of them required intervention (suturing a leaky paracentesis). Cataract progression was noted in 10 (20%) cases, which occurred in the late post-operative course (>3 months) in the majority (14%) of them, with a tube surgery- cataract surgery interval of 12.5±8.3 months. Furthermore, they reported one case of tube retraction requiring surgical revision for tube extension using 24-gauge cannular sheath, one case of persistent corneal edema requiring Descemet’s stripping automated endothelial keratoplasty (DSAEK) surgery, and 2 cases of symptomatic diplopia that were managed conservatively. The polypropylene ripcord was removed in 21 (42%) cases, at a mean of 6.9±5.6 post-operative months, resulting in significant IOP reduction (24.7±7.0 to 14.3±2.7 mmHg), with no procedure associated-hypotony (21).

Significant postoperative complications reported by Koh et al. included self-limiting shallow AC (n=7, 14.9%), hypotony requiring intervention, largely AC reformation (n=7, 9.5%), tube occlusion (n=5, 6.8%), tube exposure (n=3, 4.1%), and one case of exogenous endophthalmitis, in which Streptococcus mitis was isolated and despite device removal, loss of light perception vision ensued. Another case experienced serious complications, where persistent hypotony and subsequent plate exposure required implant removal (10).

Stent removal

The effect of removal of the ripcord or stent suture (ROS) has been reported by Mendel et al. in 145 eyes, in whom a significant IOP and medication reduction up to 3 months post-ROS was found, with a mean implantation-ROS interval of 181.9 days. The effect was notably weaker in second devices and inferiorly placed PGIs. The most commonly encountered complication was clinically significant hypotony (n=10, 6.9%), of which 6 eyes (4.1%) required surgical ripcord re-insertion. No predictive factor for post-ROS hypotony was identified (11). Chan et al. reported ROS in 43 eyes (41.7% of a PGI cohort of 103 eyes) within a minimum of 12 months of follow-up. Median implantation-ROS interval was 12.3 weeks and ROS resulted in a significant immediate IOP reduction that was strongly correlated to pre-ROS IOP values, and no hypotony was recorded as a complication (35).

PGI in special glaucoma situations

Pediatric glaucoma

Despite differing significantly from adult-onset glaucoma in other respects, childhood glaucoma is also characterized by ocular hypertension and optic disc cupping (36). Angle surgery is considered the cornerstone of surgical management of pediatric glaucoma. However, in 15–25% of cases, as with underlying structural defects, severe glaucoma, often refractory, is found at presentation (37). Of the few options available for these refractory cases, GDD have demonstrated better long-term IOP control. As children are not simply little adults, some factors should be considered when making the surgical decision of GDD implantation, including the expected disproportion between the implant size and a child’s globe size and the anticipated ocular growth over time, resulting in possible future change in device position, the very thin sclera in buphthalmic eyes, and the possible need for pars plana tube insertion, especially in aphakic eyes (38). Along with a good safety profile, valved GDDs demonstrated a good success rate in aphakic eyes and eyes with a previous failed trabeculectomy, at least in the short term, exceeding 80% at 1 year (39-41). However, the aggressive healing with the resultant late encapsulation or fibrous ingrowth within the valve leaflets—seen in children’s eyes—has resulted in a significant proportion of late failures. Therefore, non-valved GDDs may carry a higher success rate beyond 2 years (42). There is a widely-held misconception that smaller plates, such as the AGV FP8, is suitable for children. The problem is that children’s eyes scar more, not less, than adult eyes and therefore require large plates to have any chance of long-term success. So-called pediatric implants with smaller-sized plates, are, in general, rather inappropriate for high scarring eyes, whether large or small. Senthil et al. compared the outcomes of the Aurolab Aqueous Drainage Device (AADI), a novel low cost non-valved GDD that is very similar to BGI, with AGV, in 116 eyes with refractory pediatric glaucoma and found similar rates of qualified success and complication at 1 and 3 years. However, the AADI showed greater complete success, better IOP control, less need for glaucoma medications, and lesser incidence of a hypertensive phase (43).

The use of the PGI in different types of pediatric glaucoma has been reported by a number of studies. In their ongoing randomized controlled trial that compared PGI (n=25) with AGV (n=19), Elhusseiny et al. concluded comparable success (80% in PGI and 73.6% in AGV) and complications rates, and IOP and glaucoma medications reduction, between both groups at one year (44). Furthermore, Vallabh et al. reported a success rate of 84% (complete in 48%), a mean reduction of IOP and glaucoma medications at 19.1±7.7 mmHg and 2.8±1.6, respectively, over 1–2 years following PGI implantation in 25 various pediatric glaucoma cases. Regarding complications, they reported 2 hypotony cases, one of them required rip-cord re-insertion, and another case of aphakic tube blockage, in addition to one case of late corneal graft failure (45). In a study of 30 phakic eyes with refractory primary congenital glaucoma (PCG), Karapapak and Olgun also reported favorable PGI outcomes at 1 year, with a cumulative success rate of 86.6%, mean IOP and medication reduction at 24.76±9.5 mmHg and 2.7±1.3, respectively. The reported complications (n=4; 13.3%) occurred only in the first 3 post-operative months and included 3 cases of choroidal detachment and a case of tube exposure (46).

Refractory glaucoma

Although the term “refractory glaucoma” has been inconsistently used in literature, it has been most commonly employed to refer to glaucoma subtypes at high risk of failure, or cases that have already experienced failure, following conventional glaucoma filtering surgery, i.e., trabeculectomy. Traditionally, the usage of GDDs has been generally reserved for refractory glaucoma. Publications such as tube vs. trabeculectomy study (TVT), ABC, and AVB all refer to refractory glaucoma of various types. However, these large randomized studies give little indication of the usefulness of GDD in individual refractory situations. For the PGI, a number of smaller studies have focused on these areas, such as Khodeiry et al., who reported the outcomes in 30 eyes with various types of secondary refractory glaucoma (mainly neovascular glaucoma, 30% and silicone oil induced glaucoma, 30%), including a 93.3% success rate, and a mean IOP of 15.2±4.6 mmHg on 0.9±1.2 glaucoma medications, leading to a mean IOP percent reduction of 53.4%, at 1 year. The reported complications (n=4, 13.3%) were transient, except for one case that needed polypropylene ripcord re-insertion (27) Moreover, some of the studies listed in Table 4 were primarily concerned with one or more type of refractory glaucoma.

Discussion

Since its relatively recent introduction, the use of the PGI has been steadily increasing across the globe, thanks to the unique smaller tube design, which attempts to reduce the safety efficacy trade-off reported with conventional GDDs. This theoretical advantage is supported by a limited but growing body of evidence. Although currently available evidence regarding PGI outcomes is mostly retrospective and non-comparative, the evidence that exists suggests that the PGI is not inferior in efficacy over a 1–3-year postoperative time period, to the BGI.

We identified 11 studies (out of 15 listed in Table 4) that reported outcomes at 1 year. Rates of overall success exceeded 90% in the majority, with the highest rates reported by Weber et al. [2023] (95.6% in a cohort of various refractory glaucoma) (31) and Weber et al. [2025] (97% in a series of glaucoma after vitreo-retinal surgeries) (26), whereas the lowest success rate was reported by José et al. (75%) (33), a study in which a more stringent definition of surgical success was employed (18 mmHg as a cut-off and 30% IOP reduction). For the BGI, these rates were 96.1% (TVT), 86% (ABC), and 72% (AVB) after one year (47-49) Of note, complete (unmedicated) success rates varied greatly among PGI studies, essentially because these are largely retrospective, which makes inconsistencies regarding indications and timing of introducing anti-glaucoma medications expected. Mean post-operative IOP and number of medications at 1 year ranged between 11.7 mmHg on 0.4 medications by Liegl et al. (mainly POAG and pseudoexfoliation glaucoma) (24) and 15.2 mmHg using 1.7 medications by Olgun and Karapapak (neovascular glaucoma without MMC) (28). For the BGI, mean numbers after 1 year were 12.4 mmHg on 1.3 medications by TVT, 13.2 mmHg on 1.5 medications by ABC, and 13.6 mmHg on 1.2 medications by AVB.

In the few available PGI studies that reached a 3 year follow-up, overall success rates were 75% (Tan et al.; a mixed cohort of refractory glaucomas) (12) and 92% (Richardson et al.; uveitic glaucoma) (21), with mean IOP (number of medications) of 14.9 mmHg (0.2) and 12.2 mmHg (1.1), respectively. Regarding the 3-year outcomes of BGI, overall success rates were 84.9% (TVT), 67.7% (ABC), and 66% (AVB), and mean values of IOP (number of medications) were 13 mmHg (1.3), 13.1 mmHg (1.5), and 14.4 mmHg (1.1), respectively (50-52).

Unlike limbal based blebs, evaluation of blebs over equatorially placed GDDs could be challenging. Imaging techniques, including mainly ultrasonography and magnetic resonance imaging, have been used to evaluate these blebs (53,54). Using ultrasonography to evaluate PGI blebs in 70 eyes, Weber et al. identified the presence of a double bleb, which appeared to be associated with better functionality, in 78.6% of cases (55). Although bleb imaging could provide useful data about its functionality, its utility for assessing GDD surgical outcomes, including PGI, is as yet unproven.

Several complications can occur following GDD implantation (listed in Table 2). Apart from infections, serious complications associated with non-valved GDDs are mainly hypotony related, which can occur for various reasons. Hypotony may result from improper flow restriction at time of surgery, premature adjustment of a flow restrictor (sutures and/or ripcords) in response to a high postoperative IOP. Hypotony may be inconsequential (numerical hypotony) or clinically significant. The latter may result in sight threatening complications, (e.g., choroidal detachment, choroidal hemorrhage or hypotony maculopathy). Choroidal effusion was the most commonly reported post-operative complication in the first year following BGI implantation in the TVT study (16%), and equally to shallow AC in the AVB study (both =14%), whereas it came third in the ABC study (10%), after corneal edema (22%) and shallow AC (20%) within the early post-operative period (≤3 months). Reported rates of hypotony and its related complications in the previously mentioned PGI studies are similar, if not less, except for what Tan et al. reported (35.4%). As mentioned before, this particular study group employed a “stability system” technique instead of conventional flow restriction techniques. Despite this relative high rate of hypotony, they reported choroidal detachment in only 4.2% of their cases (12). Albeit this technique seems interesting, we believe that the ripcord technique provides a more predictable way of adjusting flow in different clinical situations.

Although most clinically significant hypotony typically occurs in the early post-operative period, long-term hypotony related complications may also occur. Gedde et al. reported a 4.5% cumulative rate of chronic hypotony following GDD surgery, in a pooled analysis of the TVT, ABC and AVB studies, for which BGI was the most significant predictor (hazard ratio =5.12) (56). Such a complication is sporadically reported regarding PGI and a 5-year outcomes are yet unavailable for comparison.

In the authors’ experience, the PGI dramatically reduces the wide range of early postoperative pressure fluctuation that was often seen with the BGI. This is attributed to the PGI’s smaller tube diameter, although that feature was originally included in an attempt to reduce tube exposure and corneal endothelial cell loss. The highest PGI exposure rate was reported by Weber et al. [2024] at 16.1% (9/56) within the first year. All of these cases underwent conjunctival revision along with an additional pericardial patch graft and eventually, more half of them (n=5) required device explanation despite surgical revision with a second patch graft (29). This is an unusually high rate and much higher than the authors’ experience and what was reported by most other PGI studies, including another PGI study of glaucoma after vitreo-retinal surgery conducted by the same group (26), in addition to other studies that have reported much lower rates at longer follow-up periods. At 5 years, reported BGI tube exposure rates were 5% (TVT), 2% (ABC; 1% within the first 3 months and 1% afterwards), and 2% (AVB) (57-59). Hau et al. prospectively estimated progressive corneal endothelial cell loss following BGI implantation in the AC to reach 8.1% and 36.8% centrally and 15.3% and 50.1% peripherally at 1 and 5 years, respectively (60). Comparatively, in a cohort of PGI implanted in the AC or the sulcus, Studsgaard et al. reports a mean endothelial cell loss of 4.5% at 12 months. Cell loss more than 20%, 30%, and 40% occurred in 8%, 6%, and 4% of their cases, respectively, and only one case (2%)—that had low pre-operative cell density—developed corneal decompensation (25).

Despite the relatively easier insertion that the small PGI tube might provide, it also carries a higher theoretical risk of obstruction. The highest rate of tube occlusion was reported by Tan et al. [2024] at 8.3% (n=4) cumulatively after 3 years. Half of these cases (n=2) required flushing and AC washout. One of those also required vitrectomy (12). For the BGI, rates of tube occlusion after 3 years were reported at 3% (TVT) and 4.9% (ABC).

Further to the PGI studies listed in Table 4, a fewer number of studies were primarily concerned with directly comparing PGI with another GDD, including BGI and AGV (44,61-63). They included follow-up periods for up to 12 months only and the majority of them were retrospective. In general, the PGI outcomes were at least similar and occasionally superior, especially with regards to better early IOP control, fewer number of medications and possibly lower complication rates, as summarized in Table 5.

Carlà et al. have recently conducted a systematic review that comprised 18 studies with 946 eyes and provided a moderate-to-good quality evidence to conclude that PGI may be considered in refractory glaucoma cases as an alternative to other GDDs, especially in situations when postoperative hypotony is a significant concern (64).

Limitations

This article reports the authors’ current PGI surgical implantation technique, and refers to technique variations adopted by other surgeons with regard to flow restriction. It does not provide a comparison of various technique-dependent outcomes. The evidence we provide subsequently regarding PGI is narrative in nature and based on the relatively limited available English published literature at the time of writing. This entails a finite number of studies, the majority of which are retrospective, non-comparative, having limited numbers of cases and follow-up duration and likely differing case selection.

Conclusions

The PGI is a novel GDD that is designed to minimize the well-known complications associated with non-valved implants while preserving their remarkable IOP lowering capability. From the authors’ perspective, the PGI can be employed not only to substitute the usage of conventional GDDs in special situations of glaucoma, including refractory glaucoma and pediatric glaucoma, but also to be as a primary glaucoma surgery, supported by the safety and efficacy it demonstrates. Therefore, the PGI represents a valuable tool within the glaucoma surgical armamentarium. Current evidence, albeit encouraging, is limited by various constraints and further studies addressing this knowledge gap are much needed.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Eye Science for the series “Recent Glaucoma Surgeries”. The article has undergone external peer review.

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://aes.amegroups.com/article/view/10.21037/aes-2025-1-74/rc

Peer Review File: Available at https://aes.amegroups.com/article/view/10.21037/aes-2025-1-74/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://aes.amegroups.com/article/view/10.21037/aes-2025-1-74/coif). The series “Recent Glaucoma Surgeries” was commissioned by the editorial office without any funding or sponsorship. K.B. served as the unpaid Guest Editor of the series. K.B. is a consultant for AbbVie Ltd., Advanced Ophthalmic Innovations, Alcon, Allergan, Calilia, Elios Vision, EyeD Pharma, W.L Gore & Associates Inc., iStar Medical, Ivantis Laboratoires Théa, Liquid Medical, Myra Vision, Nova Eye Medical, Ph Pharma, Radiance Therapeutics, Roche, Santen Pharmaceutical Co., Ltd., and Sight Sciences. He has received honoraria from Abbvie Ltd., Advanced Ophthalmic Innovations, Alcon, JamJoom Pharmaceuticals, Laboratoires Théa, Myra Vision, and Santen Pharmaceutical Co., Ltd. He is a stock shareholder at Aquesys, C-Mer Holdings, Ellios Vision, International Glaucoma Surgery Registry Ltd., Ivantis, MedEther Ophthalmology (Hong Kong) Ltd., Myra Vision, Vision Futures Ltd., and Vision Medical Events Ltd. He is a patent holder at the National University of Singapore. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Weinreb RN, Aung T, Medeiros FA. The pathophysiology and treatment of glaucoma: a review. JAMA 2014;311:1901-11. [Crossref] [PubMed]
2. Razeghinejad MR, Fudemberg SJ, Spaeth GL. The changing conceptual basis of trabeculectomy: a review of past and current surgical techniques. Surv Ophthalmol 2012;57:1-25. [Crossref] [PubMed]
3. Singh P, Kuldeep K, Tyagi M, et al. Glaucoma drainage devices. Journal of Clinical Ophthalmology and Research 2013;1:77-82.
4. Seibold LK. The History of Glaucoma Surgery. In: Kahook MY, Salim S, Seibold LK, editors. MIGS: advances in glaucoma surgery. CRC Press; 2024:1-11.
5. Ayyala RS, Duarte JL, Sahiner N. Glaucoma drainage devices: state of the art. Expert Rev Med Devices 2006;3:509-21. [Crossref] [PubMed]
6. Christakis PG, Zhang D, Budenz DL, et al. Five-Year Pooled Data Analysis of the Ahmed Baerveldt Comparison Study and the Ahmed Versus Baerveldt Study. Am J Ophthalmol 2017;176:118-26. [Crossref] [PubMed]
7. Shalaby WS, Reddy R, Wummer B, et al. Ahmed ClearPath vs. Baerveldt Glaucoma Implant: A Retrospective Noninferiority Comparative Study. Ophthalmol Glaucoma 2024;7:251-9. [Crossref] [PubMed]
8. Koh VT, Aquino CM, Chew P, et al. First-in-man clinical study on a novel glaucoma drainage implant for refractory glaucoma. Invest Ophthalmol Vis Sci 2019;60:3762.
9. Masdipa A, Kaidzu S, Tanito M. In Vitro Analysis of Pressure Resistance in the Paul Glaucoma Implant and Ahmed ClearPath 250 With and Without Polypropylene Thread Inside the Tube. Transl Vis Sci Technol 2025;14:2. [Crossref] [PubMed]
10. Koh V, Chew P, Triolo G, et al. Treatment Outcomes Using the PAUL Glaucoma Implant to Control Intraocular Pressure in Eyes with Refractory Glaucoma. Ophthalmol Glaucoma 2020;3:350-9. [Crossref] [PubMed]
11. Mendel L, Tan JCK, Aguilar Munoa S, et al. Outcomes of intraluminal ripcord removal from Paul glaucoma implants. Br J Ophthalmol 2025;109:1246-51. [Crossref] [PubMed]
12. Tan MCJ, Ong CW, Aquino MC, et al. Three-Year Outcomes of the Paul Glaucoma Implant for Treatment of Glaucoma. J Glaucoma 2024;33:478-85. [Crossref] [PubMed]
13. Sarkisian SR, Netland PA. Tube extender for revision of glaucoma drainage implants. J Glaucoma 2007;16:637-9. [Crossref] [PubMed]
14. Berman T, Au L. A New "Tube-in-tube" Method to Extend Glaucoma Drainage Devices Using Paul Glaucoma Implant. J Curr Glaucoma Pract 2024;18:130-3. [Crossref] [PubMed]
15. D'Alessandro E, Guidotti JM, Mansouri K, et al. XEN-augmented Baerveldt: A New Surgical Technique for Refractory Glaucoma. J Glaucoma 2017;26:e90-2. [Crossref] [PubMed]
16. Chiang MY, Camuglia JE, Khaw PT. A Novel Method of Extending Glaucoma Drainage Tube: "Tube-in-Tube" Technique. J Glaucoma 2017;26:93-5. [Crossref] [PubMed]
17. Diaz-Llopis M, Salom D, García-Delpech S, et al. Efficacy and safety of the pars plana clip in the Ahmed valve device inserted via the pars plana in patients with refractory glaucoma. Clin Ophthalmol 2010;4:411-6. [Crossref] [PubMed]
18. Luttrull JK, Avery RL, Baerveldt G, et al. Initial experience with pneumatically stented baerveldt implant modified for pars plana insertion for complicated glaucoma. Ophthalmology 2000;107:143-9; discussion 149-50. [Crossref] [PubMed]
19. Rososinski A, Wechsler D, Grigg J. Retrospective review of pars plana versus anterior chamber placement of Baerveldt glaucoma drainage device. J Glaucoma 2015;24:95-9. [Crossref] [PubMed]
20. Zhou D, Zhou XY, Mas-Ramirez AM, et al. Factors Associated with Conjunctival Erosions after Ahmed Glaucoma Valve Implantation. J Ophthalmic Vis Res 2018;13:411-8. [Crossref] [PubMed]
21. Richardson J, Tacea F, Yu J, et al. The PAUL Glaucoma Implant in the management of uveitic glaucoma-3-year follow-up. Eye (Lond) 2025;39:931-7. [Crossref] [PubMed]
22. Chhabra R, Tan SZ, Au L, et al. Long-Term Outcomes and Complications of Baerveldt Glaucoma Drainage Implants in Adults with Glaucoma Secondary to Uveitis. Ocul Immunol Inflamm 2019;27:1322-9. [Crossref] [PubMed]
23. Deubel C, Walz W, Petrak M, et al. Clinical outcomes of the PAUL Glaucoma implant for neovascular glaucoma. Graefes Arch Clin Exp Ophthalmol 2025;263:3199-209. [Crossref] [PubMed]
24. Liegl C, Bourauel L, Aretz B, et al. Clinical outcomes of the PAUL glaucoma implant in primary open-angle and pseudoexfoliative glaucoma eyes after failed glaucoma surgeries. Acta Ophthalmol 2025;104:e183-90. [Crossref] [PubMed]
25. Studsgaard A, Nielsen SE, Telinius N. One tube for all: 1-year outcomes after transition to Paul glaucoma implant at a tertiary centre. Acta Ophthalmol 2025;103:461-8. [Crossref] [PubMed]
26. Weber C, Schipper P, Walz W, et al. Clinical Outcomes of the PAUL® Glaucoma Implant for Secondary Glaucoma after Vitreoretinal Surgery. Ophthalmologica 2025;248:89-100. [Crossref] [PubMed]
27. Khodeiry MM, Hassan AK, Elhusseiny AM, et al. Outcomes of the Paul Glaucoma Implant in Refractory Secondary Glaucoma. Clin Ophthalmol 2025;19:167-74. [Crossref] [PubMed]
28. Olgun A, Karapapak M. Impact of mitomycin C on surgical outcomes of PAUL glaucoma implant in neovascular glaucoma: 12-month follow-up results. Eur J Ophthalmol 2025;35:1305-13. [Crossref] [PubMed]
29. Weber C, Hundertmark S, Stasik I, et al. Two-Year Clinical Outcomes of the PAUL Glaucoma Implant in White Patients With Refractory Glaucoma. J Glaucoma 2024;33:808-14. [Crossref] [PubMed]
30. Olgun A, Karapapak M. Assessing the Efficacy of the PAUL Glaucoma Implant in Pseudoexfoliative Glaucoma. Beyoglu Eye J 2024;9:26-32. [Crossref] [PubMed]
31. Weber C, Hundertmark S, Liegl R, et al. Clinical outcomes of the PAUL® glaucoma implant: One-year results. Clin Exp Ophthalmol 2023;51:566-76. [Crossref] [PubMed]
32. Tan MCJ, Choy HYC, Koh Teck Chang V, et al. Two-Year Outcomes of the Paul Glaucoma Implant for Treatment of Glaucoma. J Glaucoma 2022;31:449-55. [Crossref] [PubMed]
33. José P, Barão RC, Teixeira FJ, et al. One-Year Efficacy and Safety of the PAUL Glaucoma Implant Using a Standardized Surgical Protocol. J Glaucoma 2022;31:201-5. [Crossref] [PubMed]
34. Vallabh NA, Mason F, Yu JTS, et al. Surgical technique, perioperative management and early outcome data of the PAUL® glaucoma drainage device. Eye (Lond) 2022;36:1905-10. [Crossref] [PubMed]
35. Chan NS, Hing SW, Okada N, et al. Outcomes of Stent Removal in Eyes With the Paul Glaucoma Implant With Intraluminal 6-0 Polypropylene. Clin Exp Ophthalmol 2025;53:756-63. [Crossref] [PubMed]
36. Thau A, Lloyd M, Freedman S, et al. New classification system for pediatric glaucoma: implications for clinical care and a research registry. Curr Opin Ophthalmol 2018;29:385-94. [Crossref] [PubMed]
37. Weinreb RN, Grajewski AL, Papadopoulos M, et al. WGA Consensus Series 9: Childhood glaucoma. Kugler Publications; 2013.
38. Brandt JD, Hammel N, Fenerty C, et al. Glaucoma drainage devices. In: Grajewski AL, Bitrian E, Papadopoulos M, editors. Surgical Management of Childhood Glaucoma: Clinical Considerations and Techniques 2018:99-127.
39. Pakravan M, Esfandiari H, Yazdani S, et al. Clinical outcomes of Ahmed glaucoma valve implantation in pediatric glaucoma. Eur J Ophthalmol 2019;29:44-51. [Crossref] [PubMed]
40. Al-Haddad C, Al-Salem K, Ismail K, et al. Long-term outcomes of Ahmed tube implantation in pediatric glaucoma after multiple surgeries. Int Ophthalmol 2018;38:2649-52. [Crossref] [PubMed]
41. Dave P, Senthil S, Choudhari N, et al. Outcomes of Ahmed valve implant following a failed initial trabeculotomy and trabeculectomy in refractory primary congenital glaucoma. Middle East Afr J Ophthalmol 2015;22:64-8. [Crossref] [PubMed]
42. Malik R, AlDarrab A, Edward DP. Contemporary management of refractory pediatric glaucoma. Curr Opin Ophthalmol 2020;31:123-31. [Crossref] [PubMed]
43. Senthil S, Gollakota S, Ali MH, et al. Comparison of the New Low-Cost Nonvalved Glaucoma Drainage Device with Ahmed Glaucoma Valve in Refractory Pediatric Glaucoma in Indian Eyes. Ophthalmol Glaucoma 2018;1:167-74. [Crossref] [PubMed]
44. Elhusseiny AM, Khaled OM, Chauhan MZ, et al. Initial Results of the Paul Ahmed Comparison (PAC) Study in Refractory Childhood Glaucoma. Am J Ophthalmol 2025;271:71-8. [Crossref] [PubMed]
45. Vallabh NA, Mohindra R, Drysdale E, et al. The PAUL® glaucoma implant: 1-year results of a novel glaucoma drainage device in a paediatric cohort. Graefes Arch Clin Exp Ophthalmol 2023;261:2351-8. [Crossref] [PubMed]
46. Karapapak M, Olgun A. Efficacy and safety of the paul glaucoma implant in the treatment of refractory primary congenital glaucoma. Jpn J Ophthalmol 2024;68:571-7. [Crossref] [PubMed]
47. Gedde SJ, Schiffman JC, Feuer WJ, et al. Treatment outcomes in the tube versus trabeculectomy study after one year of follow-up. Am J Ophthalmol 2007;143:9-22. [Crossref] [PubMed]
48. Budenz DL, Barton K, Feuer WJ, et al. Treatment outcomes in the Ahmed Baerveldt Comparison Study after 1 year of follow-up. Ophthalmology 2011;118:443-52. [Crossref] [PubMed]
49. Christakis PG, Kalenak JW, Zurakowski D, et al. The Ahmed Versus Baerveldt study: one-year treatment outcomes. Ophthalmology 2011;118:2180-9. [Crossref] [PubMed]
50. Gedde SJ, Schiffman JC, Feuer WJ, et al. Three-year follow-up of the tube versus trabeculectomy study. Am J Ophthalmol 2009;148:670-84. [Crossref] [PubMed]
51. Barton K, Feuer WJ, Budenz DL, et al. Three-year treatment outcomes in the Ahmed Baerveldt comparison study. Ophthalmology 2014;121:1547-57.e1. [Crossref] [PubMed]
52. Christakis PG, Tsai JC, Kalenak JW, et al. The Ahmed versus Baerveldt study: three-year treatment outcomes. Ophthalmology 2013;120:2232-40. [Crossref] [PubMed]
53. Lloyd MA, Minckler DS, Heuer DK, et al. Echographic evaluation of glaucoma shunts. Ophthalmology 1993;100:919-27. [Crossref] [PubMed]
54. Iwasaki K, Kanamoto M, Takihara Y, et al. Evaluation of Bleb Fluid After Baerveldt Glaucoma Implantation Using Magnetic Resonance Imaging. Sci Rep 2017;7:11345. [Crossref] [PubMed]
55. Weber C, Weinhold L, Walz W, et al. Sonographic bleb visualisation after PAUL glaucoma implant surgery. Br J Ophthalmol 2025;109:682-8. [Crossref] [PubMed]
56. Gedde AY, Huertas L, O'Brien R, et al. Chronic Hypotony After Tube Shunt Surgery: A Pooled Analysis of Data From Three Randomized Clinical Trials. Am J Ophthalmol 2026;282:305-15. [Crossref] [PubMed]
57. Gedde SJ, Herndon LW, Brandt JD, et al. Postoperative complications in the Tube Versus Trabeculectomy (TVT) study during five years of follow-up. Am J Ophthalmol 2012;153:804-814.e1. [Crossref] [PubMed]
58. Budenz DL, Feuer WJ, Barton K, et al. Postoperative Complications in the Ahmed Baerveldt Comparison Study During Five Years of Follow-up. Am J Ophthalmol 2016;163:75-82.e3. [Crossref] [PubMed]
59. Christakis PG, Kalenak JW, Tsai JC, et al. The Ahmed Versus Baerveldt Study: Five-Year Treatment Outcomes. Ophthalmology 2016;123:2093-102. [Crossref] [PubMed]
60. Hau S, Bunce C, Barton K. Corneal Endothelial Cell Loss after Baerveldt Glaucoma Implant Surgery. Ophthalmol Glaucoma 2021;4:20-31. [Crossref] [PubMed]
61. Oliver-Gutiérrez D, Segura-Duch G, Ávila-Marrón E, et al. Paul versus Baerveldt 350 glaucoma drainage implants: One-year comparative analysis. Indian J Ophthalmol 2025;73:S317-23. [Crossref] [PubMed]
62. Berteloot S, Correia Barão R, Abegão Pinto L, et al. Treatment Outcomes Comparing the Paul and Baerveldt Glaucoma Implants After One Year of Follow-Up. J Glaucoma 2024;33:594-600. [Crossref] [PubMed]
63. Karapapak M, Olgun A. One-Year Outcomes of the PAUL Glaucoma Implant Compared With the Ahmed Glaucoma Valve for the Treatment of Silicone Oil Glaucoma. J Glaucoma 2024;33:310-6. [Crossref] [PubMed]
64. Carlà MM, Gambini G, Boselli F, et al. The Paul Glaucoma Implant: a systematic review of safety, efficacy, and emerging applications. Graefes Arch Clin Exp Ophthalmol 2025;263:2447-59. [Crossref] [PubMed]