Six-Month Preclinical Evaluation of a Bilayered ePTFE Glaucoma Implant in Normotensive Rabbits

Abstract

Purpose

The efficacy of intraocular pressure (IOP)-lowering glaucoma surgery is limited by postoperative fibrosis. A bilayered expanded polytetrafluoroethylene (ePTFE) implant reduced fibrosis compared with silicone controls in a three-month study in normotensive rabbit eyes. Here, we evaluate biocompatibility and histopathology over six months.

Methods

Twelve normotensive New Zealand white rabbits underwent bilayered ePTFE implantation in one eye and a sham procedure in the fellow eye. Clinical exams were monitored serially over six months and graded, including gross inspection, slit-lamp biomicroscopy, bleb evaluation, IOP measurement, and fundus examination. After six months, histopathologic analysis was performed on six eyes and graded for cellular integration, fibrosis, and inflammation.

Results

The implant was well tolerated. No significant differences in conjunctival hyperemia or bleb vascularity were observed between implant and sham eyes; however, tube position was closer to the cornea at six months. In implanted eyes, IOP was reduced at postoperative day 1 (12.4 ± 1.6 mm Hg, P = 0.001), week 1 (12.7 ± 1.9 mm Hg, P = 0.006), week 2 (12.9 ± 2.0 mm Hg, P = 0.002), month 1 (13.1 ± 1.5 mm Hg, P = 0.01), and month 3 (14.1 ± 2.1 mm Hg, P = 0.04) versus baseline (16.5 ± 2.9 mm Hg). Sham eyes showed reductions at day 3 and week 2. Histopathology demonstrated cellular integration into the outer ePTFE layer, with minimal inflammation (0.9 ± 1) and fibrosis (1.1 ± 0.2).

Conclusions

A bilayered ePTFE glaucoma implant was well tolerated for six months, with minimal inflammation or fibrosis.

Translational Relevance

A low-profile, bilayered ePTFE implant is well tolerated and promotes tissue integration, supporting tissue-integrated filtration as a potentially more effective glaucoma surgery.

Introduction

Filtering procedures are the primary surgical approach for patients at the highest risk of blindness from glaucoma.^1,2 These procedures lower intraocular pressure (IOP) by creating an outlet for aqueous humor (AH) drainage from the anterior chamber to the subconjunctival space. Historically, this has been achieved through bleb-based filtration, in which AH accumulates within an externally formed subconjunctival reservoir (bleb), either with implantation of a glaucoma drainage device (GDD) or without (e.g., trabeculecomy, deep sclerectomy). Examples of GDDs and subconjunctival filtration devices include the XEN Gel Stent (Abbvie Inc., North Chicago, IL, USA), PRESERFLO MicroShunt (Glaukos Corporation, Aliso Viejo, CA, USA), Ahmed Glaucoma Valve (New World Medical, Inc., Rancho Cucamonga, CA, USA), Baerveldt Glaucoma Implant (Johnson & Johnson Vision, Jacksonville, FL, USA).³ There are, however, limitations to glaucoma filtering surgeries. The need for reoperation because of inadequately controlled IOP was reported as high as 46.9% over five years after trabeculectomy and 20% to 30% after GDD placement.^4,5 Moreover, bleb-based filtration procedures are associated with a range of postoperative complications—including hypotony, bleb-related issues, implant exposure, and infection—which require intensive postoperative management, prolong patient recovery, and may limit broader adoption of these surgeries.⁶ Avoidance of these postoperative complications has contributed to a decline in the frequency of filtering surgery, with a corresponding increase in minimally invasive glaucoma surgeries.^7,8 Compared to minimally invasive glaucoma surgeries, however, filtering procedures provide greater and more sustained IOP reduction and are the recommended surgical treatment for patients with progressing advanced glaucoma.^9,10 Despite these recommendations, physicians choose to avoid filtering surgery in the majority of patients with rapidly progressing glaucoma.¹¹

Filtering surgery outcomes are strongly influenced by the postoperative wound-healing response.¹² In bleb-based systems, insufficient healing can predispose to implant exposure and chronic bleb leaks. Conversely, inadequate IOP reduction is often due to an exaggerated healing response leading to fibrosis. Thick fibrous capsules formed by a fibrotic healing response often form an impermeable barrier that prevents AH venting and, in turn, IOP reduction. Excessive postoperative fibrosis is multifactorial and results from aberrant activation of wound-healing pathways. One mechanism involves AH signaling in glaucomatous eyes, where elevated profibrotic cytokines, such as tranforming growth factor–β, promote postoperative fibrosis.^13,14 Additionally, mechanical strain from capsular wall tension within the subconjunctival space promotes capsular thickening. Increases in bleb radius of curvature lead to proportional increases in capsular wall tension and capsular wall thickness.^15–17 Last, patient-to-patient variation in healing response, the use of medications that cause chronic ocular surface toxicity, and variations in surgical technique affect post-operative healing and surgical outcomes.¹⁸ Targeting aspects of the wound-healing process is an opportunity to reduce complication rates and improve surgical outcomes.¹⁹

Attempts to target the postoperative wound healing response, however, have yielded mixed results. Application of the antifibrotic chemotherapeutic mitomycin C (MMC) at the time of surgery kills resident fibroblasts and improves surgical success after trabeculectomy.^20,21 Still, MMC application increases the risk of postoperative wound leaks and has yielded mixed results when used with GDD surgery. Use of wound-modulating biologics, such as VEGF-targeting antibodies, has shown promise but has not been widely adopted.²² Moreover, antibodies targeting the profibrotic cytokine tranforming growth factor–β failed to improve clinical outcomes over MMC in clinical trials. There is a need to explore alternative approaches to improve surgical success.^23–25

More recently, an alternative paradigm—tissue-integrated filtration (TIF)—has emerged. In contrast to bleb filtration approaches described above, which rely on the formation of a fibrous capsule to define outflow resistance, TIF uses a device-mediated reservoir and permeable interface designed to promote controlled diffusion of AH directly into surrounding tissue while encouraging biointegration. This distinction is critical, because it shifts the primary determinant of outflow resistance from a biologically variable fibrotic capsule to a more controlled, material-driven interface.

Expanded polytetrafluoroethylene (ePTFE) is a promising material in TIF glaucoma surgery. EPTFE is a biostable and biocompatible polymer that has a microstructure composed of fibers and nodes that can be tuned to interact with or exclude cells. The potential of ePTFE in glaucoma filtering surgery was first recognized decades ago and has been revisited as a component in GDD plates and trabeculectomy-modulated inserts.^26–28 Previously, a GDD plate fabricated from a bilayered ePTFE membrane containing both cell-interacting and cell-exclusion structures was shown to reduce capsular thickness compared to silicone controls and to maintain capsular permeability in rabbit studies extending to three months.²⁹ Based on these observations, tissue-integrated filtration using an ePTFE-based reservoir represents an alternative approach to subconjunctival drainage that does not rely on traditional bleb formation or fibrotic encapsulation for long-term function. Here, we report clinical outcomes and histopathological findings from a six-month rabbit study.

Material and Methods

Animals

Adult male (n = 6) and female (n = 6) New Zealand white rabbits (3.4–5 kg) were purchased (Western Oregon Rabbit Company, Philomath, OR, USA), quarantined for at least seven days before surgery, and underwent acclimation to experimental handling before use. Experimental protocols were approved by the Office of Comparative Medicine of the University of Utah. Rabbits were handled in a manner consistent with the ARVO Statement for the Use of Animals in Ophthalmic Research and the Guide for the Care and Use of Laboratory Animals.

Implant Device Materials and Construction

Test implants were fabricated from bilayered ePTFE membranes that were continuously bonded around the plate perimeter and intermittently bonded within the plate, creating a central reservoir with a nominal thickness of approximately 0.2 mm prior to implant and less than 0.5 mm when filled with aqueous (Fig. 1). The fibrils and nodes of the “open” external surface of the reservoir were designed to enable cellular integration, whereas the “tight” structure of the internal membrane surface excluded cellular structures yet permitted fluid flow. A silicone tube was sandwiched between ePTFE membranes and sealed watertight. All devices were sterilized by ethylene oxide after manufacture and packaged in sterilization pouches before implantation.

Figure 1.

ePTFE GDD design. Schematic of GDD design (A) and photo (B). Scanning electron microscopy (SEM) images of a cross-section (C), open (D), and tight (E) sides of the ePTFE membrane.

Test Device Implantation and Sham Surgery

All surgeries were performed by an experienced glaucoma surgeon (I.P.) using an operative microscope. Each rabbit underwent test implant placement in one randomly selected eye and a sham surgical procedure in the fellow eye.

Rabbits were sedated with intramuscular ketamine (50 mg/kg) and xylazine (7 mg/kg) (MWI Animal Health, Boise, ID, USA). Analgesia was provided with subcutaneous carprofen (2–4 mg/kg). Vital parameters, including pulse oximetry, heart rate, and respiratory rate, were continuously monitored during the procedure. Toe pinch was performed every 15 minutes to ensure adequate anesthesia. Supplemental ketamine was given as needed. Topical proparacaine (0.5%) and tobramycin (0.3%) were placed in the operative eye; betadine (5%) was placed in the eye and allowed to sit for one minute, and the eye was rinsed with sterile balanced saline solution before surgery. A 7-0 Vicryl stay suture was then placed in the superotemporal perilimbal cornea, and the superotemporal conjunctiva was visualized.

For test implant surgery, a 2-clock-hour fornix-based peritomy was created in the superotemporal conjunctiva, and posterior dissection was performed to create a subconjunctival pocket. The implant plate was hydrated by flushing the tube with balanced saline solution until fluid permeated the reservoir, and the implant was placed in the conjunctival pocket. A scleral tunnel was made with a 25-gauge (G) needle starting in the sclera approximately 1.5 mm posterior to the limbus, extending into the mid-anterior chamber. The implant's tube was trimmed and then placed through this scleral tract, a 10-0 nylon cross-stitch was placed across the tube entry site, and the tube plate was secured to the sclera with 10-0 nylon sutures placed through anterior eyelets. The conjunctival peritomy was reapproximated to the limbus using two 10-0 nylon wing sutures, after which proper implant and tube positioning in the anterior chamber were confirmed. Tobramycin (0.3%) and dexamethasone (0.1%) (MWI Animal Health) solution was applied to the eye at the conclusion of surgery, twice daily for seven days and once daily for seven more days. Sham surgery was performed identically to the test implant surgery up until the point of conjunctival closure. At this point, the tube and plate were removed from the eye, the scleral tract was sealed with a 10-0 nylon suture, and the conjunctiva was closed as described above.

IOP Measurement

IOP was measured with the TonoVet tonometer (TioLat, Inc., Helsinki, Finland) in awake, restrained rabbits using the magnetic probe in a horizontal orientation, according to the manufacturer's instructions. All measurements were performed by a trained technician who was masked to treatment. Before measurements, each rabbit was restrained and allowed to calm for at least three minutes. Three IOP measurements were taken of each eye; if the highest and lowest measurements differed by >3 mm Hg, the rabbit was allowed to acclimate for three more minutes before repeat measurements were taken. IOP measurements were taken as outlined in the Table on all study rabbits; all IOP measurements were performed between 9:00 AM and 12:00 PM for the duration of the study to minimize the potential influence of diurnal variation.

Table.

Rabbit Examination Schedule

			Post-Operative
Procedure	Pre-Op	Surgery	Day 1	Day 3	Week 1 (±36 Hrs)	Week 2 (±36 Hrs)	Month 1 (±3 Days)	Month 3 (±5 Days)	Month 6 (±14 Days)	Eary Withdrawal
Vet health assessment	X								X	X
Blood draws and testing	X								X	X
Weight	X	X	Updated on a weekly basis until study completion							X
Gross ocular exam	X	X	X	X	X
Slit lamp exam	X		X	X	X	X	X	X	X	X
Tonometry	X	X	X	X	X	X	X	X	X	X
Bleb evaluation			X	X	X	X	X	X	X	X
Indirect ophthalmoscopy	X			X			X		X	X
Suture removal if applicable					X	X	X	X	X	X
Adverse event assessment			X	X	X	X	X	X	X	X
Cage-side health checks			2× per day (M–F) and 1× per day (Sat–Sun)

Clinical Exams and Bleb Grading

Gross, slit lamp, dilated, and bleb exams were performed according to the schedule outlined in the Table on all study rabbits. Clinical findings on the slit lamp exam were scored using a modification of the system described by McDonald and Shadduck, with aqueous flare and cells adapted from the Standardization of Uveitis Nomenclature Working Group scale.^30,31 Implant plate position and tube position within the anterior chamber were assessed on every slit lamp examination. Scoring for the tube position within the anterior chamber was 0 = tube on the iris, 1 = tube in the mid-anterior chamber, 2 = tube in contact with corneal endothelium, and 3 = tube within the corneal stroma. Although a traditional bleb was not present, appearance of the tissue surrounding the plate was assessed and graded using a modification of the Indiana Bleb Morphology Scoring Criteria (IBAGS) that evaluates the extent, vascularity, and height of conjunctival blebs.³² Vascularity, in this system, was graded on an ordinal scale from V0 (normal vascularity) to V4 (severe vascularity). Dilated exams assessed lens opacity, vitreous haze, fundus view, and the presence of retinal perivascular sheathing. Masking was not possible for slit lamp and dilated exams. One animal was withdrawn from the study early; all available clinical examinations and IOP measurements of this animal were included in the analysis.

Histopathology

Histopathology was performed in six randomly selected eyes that had undergone test implant placement and in their fellow-eye sham-surgery controls. After euthanasia, eyes were labeled to establish orientation (e.g., a long suture placed in the lateral rectus muscle and a short suture placed in the superior rectus muscle), enucleated, and placed in Davidson's fixative, followed by 50% ethanol and 70% ethanol before processing for histopathology. After paraffin infiltration and sectioning, four slides from each implant were hematoxylin and eosin (H&E) stained and Gomori's elastin trichrome (GET) stained. All sections were interpreted by a veterinary pathologist (B.S.); cellular integration, device-related fibrosis, and device-related inflammation were graded on a semi-quantitative scale: 0 = not present; 1 = present, but minimal feature; 2 = notable feature, mild; 3 = prominent feature which may disrupt tissue architecture but is not overwhelming, moderate; 4 = overwhelming feature or feature that effaces or disrupts tissue architecture, severe.

Statistical Analysis

Sample size was based on prior rabbit glaucoma drainage device studies and historical experience with similar implant evaluations. A formal a priori power calculation was not performed.^33,34 All results are displayed as mean ± SD. For IOP, bleb vascularity, and conjunctival hyperemia, two-way analysis of variance (ANOVA) with Dunnett's multiple comparisons test was used. For bleb height and tube position, a one-way ANOVA with Tukey's post hoc test was used. P-values or multiplicity-adjusted P-values were calculated, and values <0.05 were considered significant.

Results

Clinical Observations

No signs of device-related animal stress were observed throughout the study. Device implantation did not affect body weights or general, non-ocular health. One rabbit failed to gain weight and exhibited reduced fecal output at postoperative week 6. The rabbit met criteria for withdrawal at postoperative week 10 because of reduced weight and was euthanized, and a necroscopy was performed. On necroscopy, the left lower molar was observed to be growing into the tongue, and the tongue exhibited an extensive area of ulceration with necrosis and marked neutrophilic infiltrates. The cause of weight loss and inappetence in this rabbit was consistent with dental malocclusion syndrome. Gross exam results of all test and control eyes were normal throughout the study without signs of infection, inflammation, hemorrhage, or wound leaks.

Slit Lamp Exams and Bleb Grading

Slit lamp images of test eyes are shown in Figure 2. Exams revealed mild conjunctival hyperemia, swelling, and discharge in both test and control eyes, which resolved by the postoperative month-1 exam. There were no significant differences in conjunctival hyperemia between test and sham eyes at any time point (Fig. 3). In the first post-operative week, anterior chamber blood was noted in two test eyes at the site of tube entry, and one sham-operated eye, where the needle entered the eye to create the scleral tract. Anterior chamber blood clots in test and sham eyes were resolved by post-operative weeks one and two, respectively. Fibrin was noted in test eyes (three of 12 eyes) on postoperative day 1 and was resolved on the post-operative day 3 exam.

Figure 2.

Clinical images. Images of the ePTFE GDD immediately after (A), one week (B), one month (C), three months (D), and six months after placement (E). Arrowhead points to the tube in the anterior chamber, and asterisk denotes the region of the ePTFE GDD plate.

Figure 3.

Clinical grading. (A) There were no significant differences in conjunctival hyperemia between Gore ePTFE- and sham-operated eyes. (B) Bleb Height was stable throughout the study (n = 12). There were no significant differences in bleb vascularity (C) between Gore ePTFE- and sham-operated eyes. POD, postoperative day; POM, postoperative month; POW, postoperative week.

Bleb heights in test eyes varied from mild to moderate elevation over the course of the study, and mild to moderate vascularity resolved to normal vascularity on the postoperative month 3 exam (Fig. 3). In sham eyes, blebs did not form (IBAGS graded height was 0 for the duration of the study); however, vascularity was noted as mild in 10 eyes through the postoperative day 14 slit-lamp exam. Bleb vascularity scores (IBAGS grade) between test and sham eyes were not significantly different at any time point (Fig. 3).

Implant position remained stable throughout the study, with no plate migration, conjunctival retraction, or implant exposure. Tube position within the anterior chamber was noted and in the mid-anterior chamber for most eyes, with some tubes noted on the surface of the iris and others noted adjacent to the corneal endothelium. There was anterior migration towards the corneal endothelium of tubes within the anterior chamber at the postoperative month 6 exam (Fig. 4). The tube position was graded more anteriorly on the postoperative month 6 exam with a mean score of 1.5 ± 0.5, which was significantly greater than postoperative day three (0.8 ± 0.5, P = 0.01, one-way ANOVA), week one (0.8 ± 0.5, P = 0.005), week two (0.8 ± 0.5, P = 0.04), and month one (0.8 ± 0.4, P = 0.04).

Figure 4.

Tube position. Schematic (A) of the anterior chamber and grading system for tube position. (B) The anterior chamber tube location was significantly more anterior at six months than at baseline (n = 11).

Dilated exams were unremarkable throughout the study. No lens opacities were observed, and posterior segments were normal in test and sham-operated eyes.

IOP Measurements

Mean preoperative IOP of test eyes at baseline was 16.5 ± 2.9 mm Hg. Mean preoperative IOP of sham eyes at baseline was 15.5 ± 2.8 mm Hg (P = 0.4, t-test). IOPs were within the physiological range across rabbits and time points, ranging from 10 to 24 mm Hg (Fig. 5). Test eye IOPs were reduced on postoperative days one (12.4 ± 1.6 mm Hg, P = 0.001, two-way ANOVA), week one (12.7 ± 1.9 mm Hg, P = 0.006), week two (12.9 ± 2.0, P = 0.002), month one (13.1 ± 1.5, P = 0.01), and month three (14.1 ± 2.1, P = 0.04) compared to baseline preoperative IOP. This reduction was not present at month 6 (P = 0.9). For sham-operated eyes, a significant reduction was observed between baseline and postoperative day three (13.1 ± 2.1 mm Hg, P = 0.02) and week two (12.6 ± 1.1 mm Hg, P = 0.005). Otherwise, no significant reductions in IOP were noted in sham-operated eyes.

Figure 5.

IOP measurements. IOP in test eyes was reduced at POD1, POW1, POW2, POM1, and POM3 versus baseline. IOP in sham-operated eyes was reduced at POD3 and POW2 versus baseline (n = 12).

Histopathology

Histopathologic examination of test eyes was significant for cellular integration into the outer layer of the implant, cell exclusion by the inner layer, and minimal device-related tissue response (Fig. 6). On examination of H&E- and GET-stained sections, the outer layers of each implant exhibited complete integration (score = 4.0 ± 0, n = 6), with no cell infiltration through the inner tight layers. No cellular infiltration was observed in the device lumens or at the junctions. Device-related inflammation (score = 0.9 ± 1, n = 6) and fibrosis (score = 1.1 ± 0.2, n = 6) scores were each minimal. The surgical sites of the sham-operated control eyes were characterized by minimal fibrotic healing response and little to no evidence of inflammation.

Figure 6.

GDD histopathology. H&E (A–C) and GET (D–F) staining GDD implants 6 months after placement. Low magnification images of the anterior portion of the GDD (A, D) with tube lumen (asterisk) and a section through the GDD plate (B, E) showing inflated reservoir lumen with lightly basophilic material (arrow) consistent with aqueous humor retention. Higher-magnification images of tube plate (C, F) demonstrating cellular integration into the outer layer (arrow) and exclusion from the tube lumen by the inner tight layer (asterisk).

Discussion

Here, we present data on the long-term biocompatibility of a GDD fabricated from dual-layered ePTFE using a normotensive rabbit model, which is considered the standard for preclinical evaluation of subconjunctival GDD. Surgeries were well tolerated, and animals recovered rapidly without any device-related complications. Scores for ocular surface reaction(s) in test eyes returned to baseline after the first postoperative month and largely mirrored those in eyes that underwent sham procedures. Anterior chamber bleeding and fibrin reactions resolved within one to two weeks in test and sham-operated eyes as well. EPTFE is a stable and inert polymer of tetrafluoroethylene with a track record of biocompatibility and success as a component of biomedical implants. It has a highly tunable microarchitecture, and its flexibility can be tailored for various applications, including sutures, vascular grafts, and surgical meshes for hernia repair.³⁵ Here, we extend on previous results by demonstrating the biocompatibility of a novel bilayered ePFTE in the subconjunctival space of rabbit eyes over six months.

Our findings here build on prior studies that highlighted ePTFE's promise in glaucoma filtering surgery. EPTFE was initially investigated as an aqueous wick and as a covering for a GDD-like device in the late 1980s, demonstrating excellent biocompatibility and encouraging clinical results.^26,27 Ahn et al.²⁸ subsequently validated this approach clinically in patients with refractory glaucoma, reporting IOP reduction in nine of eleven patients over 13 months of follow-up. Kim et al.²⁷ reported a 65% overall success rate and a Kaplan-Meier IOP control survival of 60% at three years among 40 patients who received a 160 mm² ePTFE plate.^33,34 Although ePTFE-based GDD did not reach widespread clinical adoption following these early studies, they established a foundation for continued investigation of ePTFE as a platform to improve glaucoma surgical outcomes.

Cellular integration is a central and intentional design feature of this implant, intended to reduce the biologic drive to form a thick, fluid-impermeable capsule. The bilayered architecture distinguishes this device from prior ePTFE designs: whereas a denucleated, large-pore ePTFE implant previously shown to reduce capsular thickness to 30% of that surrounding Baerveldt implants relied on cellular integration alone,³⁵ the current design pairs an open outer layer that actively promotes cellular ingrowth with a tight inner layer that permits aqueous flow while excluding cellular invasion of the device lumen. Our histopathologic findings at six months directly support this mechanistic rationale. The outer ePTFE layer achieved complete cellular integration (score 4.0 ± 0) in all examined eyes, while the inner tight layer excluded cells from the device lumen. Device-related inflammation and fibrosis scores were minimal (0.9 ± 1 and 1.1 ± 0.2, respectively). These findings are consistent with the hypothesis underlying this device's design—and with our prior demonstration of reduced capsular thickness compared to a silicone control at three months²⁸—whereby cells integrating into the outer membrane share the mechanical load with the ePTFE scaffold, reducing the biomechanical stimulus for capsular thickening.²⁸ Beyond load-sharing, micro- and nano-structured surface topographies have independently shown to reduce fibrosis biomarker expression and capsular thickness,³⁶ suggesting that the tunable microarchitecture of ePTFE may modulate the healing response through multiple complementary pathways.

This approach reflects an evolving conceptual framework for antifibrotic biomaterials in glaucoma surgery. Promoting tissue integration into the GDD contrasts with traditional glaucoma filtering approaches that favor inert materials that do not interact with surrounding cells. While inert materials do not promote fibrosis on their own, the high frequency of traditional GDD surgical failures demonstrates that they do not prevent postoperative fibrosis in many cases. We propose that an antifibrotic material is not simply a passive barrier to wound healing, but rather a surface or scaffold designed to modulate the cellular and molecular processes that drive fibrosis. Although early approaches emphasized inertness to minimize tissue reaction, it is now well recognized that completely inert materials often provoke dense, avascular encapsulation.³⁷ In contrast, effective antifibrotic materials typically engage surrounding cells in a controlled manner—regulating fibroblast adhesion, migration, mechanotransduction, and phenotype to promote organized tissue integration rather than scar formation. The bilayered ePTFE design operationalizes this principle: the open outer layer recruits and integrates host cells in a structured scaffold, whereas the tight inner layer maintains a cell-free fluid conduit. By tuning the microarchitecture of ePTFE, this approach biases the interface toward thinness and increased porosity, allowing continued aqueous outflow.

The IOP data from this study warrant careful interpretation in the context of the normotensive rabbit model. Test eyes demonstrated significant IOP reductions from baseline through postoperative month 3, after which IOP returned to baseline levels by month 6. Sham-operated eyes also exhibited transient IOP reductions at postoperative day 3 and week 2, likely reflecting the physiologic effects of the sham surgery, which included creation of a tunnel from the anterior chamber to the subconjunctival space. Because of their low baseline pressures and vigorous scarring response, rabbits are a challenging model for sustained IOP endpoints: trabeculectomies in New Zealand White rabbits fail within 10–14 days without MMC and extend to approximately 40 days with its use.^37,38 Bleb survival rather than durable IOP reduction is the conventional measure of success in that setting. Rabbit GDD studies accordingly rely on capsular thickness, histologic outcomes, and flow resistance at predefined endpoints—features we previously evaluated and found to favor the bilayered ePTFE design over a silicone control.²⁸ Against this backdrop, the normalization of IOP at month 6 in test eyes is consistent with progressive and aggressive capsular maturation inherent to the rabbit model. These results should therefore not be interpreted as evidence that the implant cannot provide sustained aqueous humor drainage in glaucomatous eyes. Instead, they highlight the limitations of the normotensive rabbit model for predicting long-term clinical efficacy, which will require evaluation in models with elevated IOP and clinically representative wound-healing response.

Anterior displacement of the intracameral tube was observed at the six-month examination despite stable positioning of the implant plate. This finding is clinically important because tube proximity to the corneal endothelium may increase the risk of endothelial injury over time. The mechanisms underlying this observation cannot be determined definitively from the present study but are likely multifactorial and model-specific. New Zealand rabbit eyes grow rapidly during the first year of life with significant increases in axial length, scleral thickness, and have accompanying corneal flattening.^38,39 Therefore, in a device with a fixed plate location, it would be expected that tube position could change over time. Additionally, the mechanical properties or remodeling of the rabbit sclera could allow the tube to straighten over time. This mechanism was proposed as a cause of anterior tube migration towards the cornea in pediatric glaucoma cases.⁴⁰ Although no eyes exhibited corneal endothelial touch or complications during the study period, progressive displacement over longer durations could raise concern for endothelial cell loss and corneal decompensation. As with all anterior chamber GDD, tube position should be optimized during placement and closely monitored to reduce the risk of corneal decompensation.

These studies have several recognized limitations. The sample size is appropriate for assessing preclinical biocompatibility; however, it limits statistical power to detect smaller effect sizes across additional outcomes. We used sham surgery as the comparative arm rather than an active GDD. Although this design isolated the tissue response attributable to implantation of the ePTFE device itself, inclusion of a clinically used GDD or the silicone control used in our previous study would have provided a more clinically relevant benchmark for device performance and reduced the potential for observer bias during clinical examinations. The present study was designed primarily to evaluate long-term safety, biocompatibility, and histopathologic integration under GLP conditions, whereas comparative efficacy against existing devices has been addressed previously for silicone controls and will be an important focus of future studies. An additional limitation of this study is that outflow facility and distal aqueous flow parameters were not assessed. Although these measurements would provide additional insight into the functional performance of the implant, the primary objective of this study was the long-term evaluation of safety, biocompatibility, and tissue integration. To preserve tissue integrity for these objectives, outflow was not assessed. Last, this study was performed in normotensive animals, which do not replicate the elevated IOP, altered aqueous humor composition, or postoperative wound-healing environment present in glaucomatous eyes. Consequently, although this model is well established for assessing long-term device biocompatibility, it cannot fully evaluate the implant's antifibrotic or long-term filtration performance under the biochemical and mechanical conditions encountered in glaucoma. Future studies in ocular hypertensive or glaucomatous eyes will be important to further assess the device's long-term therapeutic performance.

Conclusions

In this six-month preclinical study, a bilayered ePTFE glaucoma drainage implant demonstrated excellent durable biocompatibility and was well tolerated in normotensive rabbit eyes. There were no device-related complications, and clinical findings in implanted eyes were comparable to those of sham-operated controls. Histopathologic examination revealed complete cellular integration into the outer ePTFE membrane layer, with minimal inflammation and fibrosis. IOP reduction persisted through three months postoperatively in device eyes. These findings extend prior work demonstrating reduced capsular thickness with this membrane design and support continued development of ePTFE-based TIF as a strategy to mitigate postoperative fibrosis.