Rabbit Study of an In Situ Forming Hydrogel Vitreous Substitute

Abstract

purpose. An in situ forming hydrogel was evaluated as a potential vitreous substitute in rabbits.

methods. The hydrogel used a disulfide cross-linker that was then reduced to produce an injectable thiol-containing polymer solution. The disulfide cross-links reformed by air oxidation of the thiols and produced a stable hydrogel once inside the eye. The polymer was clear, autoclavable, and could be stored easily in the presence of nitrogen gas. Capillary rheometry was used to measure the viscoelastic properties of the hydrogels and the porcine vitreous. Fourteen black rabbits underwent a pars plana, 25-gauge, three-port vitrectomy by a single surgeon with injection of a vitreous substitute.

results. The refractive indices of the hydrogels were measured by refractometry and were shown to be close to 1.33, and the 2% hydrogel matched the mechanical properties of the natural vitreous humor. The reduced polymeric hydrogel was easily injectable through a small-gauge needle into the vitreous cavity and did not show any fragmentation. The material underwent gelation within the eye, remained optically clear, and appeared well tolerated clinically. Slit lamp examination, dilated fundus examination, and electroretinograms showed no evidence of vitritis, uveitis, or endophthalmitis after 1 week. Histopathologic evaluation did not reveal any overt toxicity or gross morphologic changes in the retina.

conclusions. The fact that this process of in situ gelation gives rise to hydrogels that are biocompatible and physically and optically similar to the natural vitreous suggests its suitability as a permanent vitreous substitute. Hydrogel candidates will be further studied to evaluate long-term biocompatibility and degradation in vivo.

Occupying two thirds of the volume of the eye, the vitreous is vital to the eye in terms of its development, mechanical stability, biochemical transport, and optical clarity. Although the vitreous consists almost exclusively of water, its exact composition has been rigorously studied for centuries. ¹ The vitreous is a hydrated gel containing greater than 98% water containing an unbranched collagen fibril network interspersed with the glycosaminoglycan, hyaluronan. The collagen is heterogeneously distributed and comprises types II, V, VI, IX, and XI. ² Sebag and Balazs ³ correlated macroscopic structure with the ultrastructural findings to show that the adult vitreous is composed of aggregate parallel collagen fibrils without the existence of any membranous structures. 

The vitreous and its interactions within the eye become important in the context of many clinically encountered vitreoretinal diseases such as posterior vitreous detachment, retinal detachment, vitreous hemorrhage, and uveitis. In an era of increasing rates of primary vitrectomies, there is a need for improved vitreous substitutes, especially in the context of retinal detachment where mechanical endotamponade is essential. Currently used vitreous substitutes include silicone oil, perfluorocarbons, air, and gases such as sulfur hexafluoride and perfluoropropane. Although many of these are useful for short-term endotamponade, researchers have yet to produce a successful long-term or “permanent” vitreous substitute. Many natural and semisynthetic compounds have been evaluated for this purpose, but have been discarded due to early degradation in these in vivo studies. ⁴  

The use of synthetic polymers began in the 1950s, shortly after the introduction of poly(methyl methacrylate) as an intraocular lens. ⁵ Since then, many polymers have been examined and proposed including poly(vinyl alcohol), poly(1-vinyl-2-pyrrolidone), poly(acrylamide), poly(glyceryl methacrylate), poly(methyl-2-acrylamido-2-methoxyacetate), and poly(2-hydroxyethylacrylate), and these have been reviewed by several groups. ^{6 7 8} Previously developed hydrogels have not been suitable because of retinal toxicity, biocompatibility, increased intraocular pressure, and the formation of opacities. ⁵ In addition, previous work with preformed polymeric hydrogels has demonstrated problems with fragmentation as well as changes in the viscoelastic properties and resiliency after injection through a small-gauge needle, ^{5 9} or complications associated with the trauma of implantation of a preformed hydrogel. ¹⁰  

Muller-Jensen and Kohler ¹¹ initially investigated poly(acrylamide) as a vitreous replacement hydrogel and early experiments showed a toxic uveitis. Their later studies with a different cross-linking agent showed improved biocompatibility and optical clarity up to 14 months, suggesting its possibility as a vitreous substitute. ^{5 12}  

The use of poly(acrylamide) as an injectable hydrogel is not novel in itself, as it was commonly used in Eastern Europe and China in the 1990s by plastic surgeons for breast augmentation. ¹³ Based on our review of the literature, it appears that we are the first to develop a thiol-containing, in vivo forming hydrogel using copoly(acrylamide). We chose acrylamide as a readily available monomer with known neurotoxicity to show that our process of rigorous purification generates a monomer-free biocompatible polymer. Our group previously prepared and described a prototype hydrogel with pendant thiol groups, as a proof of concept, which was shown to undergo endocapsular gelation for the development of an injectable accommodating intraocular lens. ¹⁴  

In this article, we extend our proof of concept and report the preparation of a similar copolymeric hydrogel that can be injected via a small-gauge needle in liquid form and undergo gelation within the vitreous cavity in the presence of physiologic oxygen concentration. ^{15 16} Further, the mechanical and optical characteristics of the hydrogel were extensively evaluated before their use in vivo in an attempt to match their properties to that of the natural vitreous humor. Last, we show that this concept of in vivo gelation is a practical consideration for vitreous substitutes, as it maintains biocompatibility and optical clarity in the rabbit model. 

Methods

Vitreous Substitute Preparation

Materials.

Acrylamide (AAm) (Bio-Rad, Richmond, CA), N-phenylacrylamide (NPA), bisacryloylcystamine (BAC), N,N,N′,N′-tetramethylethylenediamine (TEMED), dithiothreitol (DTT), Dulbecco’s phosphate-buffered saline (DPBS; Sigma-Aldrich, St. Louis, MO), and ammonium persulfate (APS; Aldrich, Milwaukee, WI) were purchased and used without further purification. All other reagents used were of analytical grade. 

Synthesis of Copolymers.

The copolymers were prepared at acrylic mole ratios of 92.5% AAm, 4.5% BAC, and 3.0% NPA. The copolymeric hydrogels were synthesized in 25% ethanol/water (wt/wt) at an initial monomer weight concentration of 7.5%. The hydrogels were formed by free-radical polymerization with 10% APS and TEMED as described in previous work. ¹⁶  

Reductive Liquefaction and Purification.

After exhaustive washing with double distilled water to remove residual unreacted monomer, DTT was used to reduce the hydrogels from the S-S cross-links to linear polymers with S-H groups. To improve the biocompatibility and decrease polydispersity, the reduced polymer solutions were dialyzed with tubing (Spectra/Por; Spectrum Laboratories, Rancho Dominguez, CA) with a 25-kDa molecular mass cutoff in pH 4 nitrogen-bubbled water, to prevent oxidation. The dialysis washing solution was changed three times over 72 hours. Subsequently, the polymer was precipitated in pH 4, nitrogen-bubbled methanol. The precipitate was lyophilized, producing a sterile sample, and was stored under vacuum to prevent oxidation. 

Characterization of Soluble Copolymers.

The molecular masses of the reduced water-soluble copolymers were determined using gel permeation chromatography (GPC; Viscotek, Houston, TX) equipped with static light scattering, refractive index, and viscosity detectors in tandem. Poly(ethylene glycol) samples of known molar masses were used to standardize the column. 

Regelation.

The reduced polymers will regel after an hour by air oxidation at physiological pH and 5% oxygen concentration. ¹⁶ Figure 1shows the complete reaction scheme of the acrylamide copolymer, and the reaction mechanism for oxidative regelation of the copolymers with air can be found in a previous publication. ¹⁴ Polymers were prepared at concentrations of 2% and 3% (wt/wt) for regelation in dialysis tubing. Regelation in dialysis tubing in pH 7.4 1× DPBS was used to simulate the physiologic conditions of the eye by enabling the diffusion of oxygen. 

Viscoelastic Characterization of Porcine Vitreous Humor and Hydrogels.

Viscoelastic properties of the porcine vitreous and hydrogels were determined with an oscillatory capillary rheometer (Vilastic-3; Vilastic, Austin, TX). The capillary rheometer was used for rheological evaluation because it enabled testing of small samples and the central section of the porcine vitreous rather than the encapsulated vitreous body which would show slippage effects in parallel plate rheometry. ¹⁷ The capillary tube had an inner diameter of 0.0991 cm. The anterior segment was removed from fresh porcine eyes which were obtained from a local abattoir (Weyhaupt, Belleville, IL). The vitreous was removed from the vitreous cavity, and 0.4 cm³ of the intact vitreous was aspirated for testing. All vitreous samples were evaluated at 25°C at 2 Hz, with increasing shear rate. Vitreous samples from 87 porcine eyes were tested individually, and the results can be found in a previous publication. ¹⁶ Porcine vitreous samples were tested at 25°C rather than at physiological temperature because Tokita et al. ¹⁸ showed the mechanical properties of the vitreous were temperature invariant. The 2% and 3% hydrogel samples were tested in triplicate at 37°C and 2 Hz, with increasing shear rates. 

Refractive Index.

A refractometer was used to measure the refractive index of each hydrogel sample (ATAGO Refractometer NAR-1T; Abbe, Kirkland, WA). The testing was done at a visible light wavelength of 552 nm at 37°C. 

Animal Preparation

Fourteen black rabbits were used for purposes of the in vivo studies involving vitrectomy, injection of the developed vitreous substitute, and subsequent testing of toxicity. Animals were cared for and handled in accordance with the ARVO Statement on the Use of Animals in Vision and Ophthalmic Research and in accordance with institutionally approved protocols. All surgeries were performed with sterile technique as used routinely under the standard of care for ophthalmic surgery in human subjects. The technique included a 5% povidone/iodine preparation of the periocular tissues and instillation of dilute povidone onto the conjunctival tissues. The rabbits were appropriately positioned and fixed on an operating table. Skin electrodes were applied to monitor physiological parameters. Body temperature was maintained by a cushion of heated air. After pupillary dilatation was established, the rabbit was placed into a restrainer and administered ketamine (15 mg/kg) and acepromazine (1 mg/kg) intravenously for short-term anesthesia. No preoperative antibiotic regimen was used. Atropine (1%) and triple-antibiotic ophthalmic ointment were instilled under the eyelids of the surgical eyes. All animals had vital signs monitored on a heating pad until they regained full consciousness. A summary is shown in Table 1 . 

Vitrectomy and Injection of Vitreous Substitute

All surgeries were performed under a standard ophthalmic operating microscope (Carl Zeiss Meditec, Inc., Oberkochen, Germany). A standard three-port, 25-gauge, sutureless trocar–cannula vitrectomy system was used (Alcon Laboratories, Fort Worth, TX). Initially, a 25-gauge trocar was placed in the infratemporal quadrant (∼7 o’clock) 3.5 mm posterior to the limbus to which an infusion line was connected. Two further trocars were similarly placed through the pars plana in the two superior quadrants, allowing for the introduction of a light pipe and vitrector. A core vitrectomy was performed and a vitreous detachment created, followed by a gentle vitrectomy of the peripheral vitreous base. At the end of the procedure, the infusion cannula and trocar were removed, and the reduced hydrogel was then injected via a 25-gauge needle through a superior trocar, until egress was witnessed from the fellow trocar. At that point both remaining trocars were withdrawn, and the overlying conjunctiva massaged. There were no leaks from the sclerotomies and no sutures were necessary. One rabbit developed an iatrogenic hole in the peripheral retina during vitrectomy which was included and followed in the study. 

Surgical Protocol

Fourteen rabbits underwent vitrectomy of the right eye with injection of different vitreous substitutes (2% hydrogel, n= 9; 3% hydrogel, n = 1; air, n = 4). The left eye served as the control in all animals and was harvested at the time the right eye was enucleated. Nine rabbits underwent vitrectomy with placement of a 2% hydrogel vitreous substitute in the right eye, of which five were harvested on postoperative day 1 and four on postoperative day 7. One rabbit underwent vitrectomy with placement of 3% hydrogel vitreous substitute, and its eyes were enucleated on postoperative day 7. The 3% hydrogel was included in the study to determine the feasibility of its injection into the vitreous cavity. The eyes of four rabbits were vitrectomized and received an air endotamponade, and the eyes of two of them were enucleated on day 1 and two on postoperative day 7. Fundus examinations of all vitrectomized eyes were performed before vitrectomy (baseline), on postoperative day 1, and at day 7 for those that were observed for 1 week. Electroretinogram (ERG) testing with a standardized protocol was performed on six rabbits with eyes enucleated on postoperative day 7 (n = 6). However, the ERG data of only five rabbits was used for analysis, excluding the one with an iatrogenic hole and subsequent detachment. 

Examination

Slit Lamp.

An ophthalmic table slit lamp (Haag-Streit, Bern, Switzerland) was used to perform examinations at baseline, 1 day after placement of a vitreous substitute, and again at postoperative day 7 in those animals that were enucleated at 7 days. The left eye was also examined and slit lamp photos were taken during examination for documentation. 

Intraocular Pressure.

Intraocular pressure (IOP) was estimated by digital palpation of the globes. This test was performed on both eyes by a single examiner at baseline and postoperative days 1 and 7. 

Electroretinograms.

The ERGs were performed on six rabbits on postoperative day 7. All subjects received 1 gtt (drop) of tropicamide and 1 gtt of 2.5% phenylephrine for mydriasis after undergoing sufficient dark adaptation (∼45 minutes). Testing was performed with an ERG system (Espion E² with ColorDome; Diagnosys LLC, Littleton, MA) in rabbits under general anesthesia. An infant-sized electrode (Burian-Allen) was placed onto the cornea. The modified ISCEV standard protocol was used in all cases. Assurance was made to keep the eye position and resistance in the ColorDome balanced between the two eyes. The ERG amplitudes were recorded, and time constants of 0.01, 0.1, and 1 second were used. The a-wave amplitude was defined as the difference in amplitude between the baseline to the trough of the a-wave. The b-wave amplitude was defined as the distance from the bottom of the a-wave to the peak of the b-wave. 

The data were analyzed to compare the right eyes with their respective left eye controls. Furthermore, the ratio of right to left amplitudes in the subset receiving the vitreous substitute versus the air endotamponade was compared. Although raw data were obtained from six rabbits, analysis of data was performed on only five rabbits, as the one with an iatrogenic retinal detachment was excluded. 

Pathology.

A subgroup of seven rabbits was euthanatized on postoperative day 1 and another group of seven on postoperative day 7. Both the right (experimental) and left (control) eyes were harvested from these animals. They were removed and fixed in 4% paraformaldehyde solution. After gross examination, the anterior segment was dissected, and the posterior segment cups were embedded in paraffin for only those specimens observed for 1 week (2% hydrogel, n = 4; air, n = 2). Subsequently, three consecutive sections were obtained of each eye and stained with hematoxylin-eosin (H&E) by a veterinary histologist. These sections were officially reviewed by a pathologist. 

Results

In Vitro

The poly(acrylamide) copolymer hydrogel was reproducibly synthesized, reduced, and regelled in the laboratory. In vitro testing demonstrated that the developed compound was reproducibly regelled by air oxidation at 2% to 3% polymeric concentration in physiologic saline solution. Gelation time within the dialysis tubing in the saline solution at oxygen concentrations of 5% and 20% measured 60 and 30 minutes, respectively. The refractive indices of both the 2% and 3% formulations were measured to be 1.338 by use of a refractometer. The number average molecular mass of the copolymer was measured to be 207.2 kDa by GPC, with no polymer chains smaller than 10 kDa. The polydispersity index of the copolymer was 2.36. In addition, the molecular mass between cross-links was determined to be 3.5 kDa. 

Capillary rheometry was used to compare the viscoelastic properties of the vitreous substitute to those of the porcine vitreous humor. A viscoelastic material exhibits solid-like and liquid-like behavior. The storage modulus (G′) represents the elastic or solid-like component, while the loss modulus (G″) represents the viscous or liquid-like component. The storage modulus is slightly higher than the loss modulus in the vitreous humor, which indicates that the vitreous behaves as a viscoelastic solid. Therefore, the vitreous substitutes should also have a slightly higher storage than loss modulus. The porcine vitreous humor has a storage modulus of approximately 5 Pa and a loss modulus of 2 Pa at a frequency of 2 Hz. Figure 2shows that the viscoelastic properties of the vitreous substitutes closely matched those of the natural vitreous humor. 

In Vivo

Slit lamp examinations revealed no significant inflammation or other disease of the anterior segment at examinations on postoperative days 1 and 7. Trace cells were seen in all surgical eyes on postoperative day 1, which resolved by the examination on day 7. In two of the rabbits, a vacuole was present in the anterior vitreous cavity that appeared to be consistent with an air bubble, as shown in Figure 3 . On digital palpation of the globes, none of the rabbits presented with unusually firm eyes at the examinations on either 1 or 7 days after surgery. In fact, the right eyes felt comparable to their counterpart left eyes. 

Serial funduscopic examination of the tested eyes revealed no evidence of vitritis, uveitis, retinitis, or endophthalmitis (Figs. 4 5) . One rabbit demonstrated a retinal hole which was iatrogenic in nature and developed into a localized detachment by the examination on postoperative day 7. In the remainder of the specimens, there was no evidence of retinal detachments or subretinal fluid. 

In the subgroup receiving the hydrogel vitreous substitutes, there was no statistically significant difference in the ERG amplitudes of low, medium, or high b-waves in comparison to the left eye controls at postoperative day 7. Although there were small absolute differences in a-wave and flicker responses, they were not statistically significant (P = 0.13 and P = 0.12, respectively), as shown in Figure 6 . The air endotamponade group results also demonstrated these small differences in the a-wave and flicker response amplitudes, as shown in Figure 7 . Comparison of the air endotamponade subgroup with the vitreous substitute subgroup by way of right/left eye amplitude ratios yielded ERG results similar to comparison of the vitreous substitute group versus the control (Fig. 8) . 

Examination of H&E stained retinal sections under the light microscope revealed that the integrity of the retinal layers and RPE appeared preserved with no evidence of toxicity or vacuolization. A few specimens demonstrated inflammatory cells localized only to the preretinal space, primarily near the vitrectomy site with some spillover posteriorly. Sections of the right (experimental) eyes that received the 2% hydrogel vitreous substitute were otherwise identical by light microscopy to those of the respective left eyes (Fig. 9) . 

Discussion

The search for an ideal vitreous substitute has been an elusive task, as many biopolymers and synthetic compounds have been examined experimentally over the past century. Natural polymers such as collagen and hyaluronic acid, and semisynthetic polymers such as hydroxypropyl methylcellulose (HPMC), fell short of being suitable vitreous substitutes because of quick biodegradation, as witnessed in several in vivo studies. ^{4 19} Rather, these substances may be better suited as intraoperative materials and drug delivery scaffolds. Poly(vinylpyrrolidone) (PVP) was the first synthetic polymer to be experimentally injected as a vitreous substitute in 1954. ⁵ Because of its reasonable biocompatibility, there has been a resurgence in interest in PVP; however, the primary problem with PVP is that it is injected as a preformed hydrogel which results in opacities and hydrogel fragmentation in vivo. ^{20 21}  

Intravitreal silicone oil, which was initially evaluated in 1962, is the most widely used synthetic, long-term vitreous substitute. ²² Unfortunately, keratopathy, glaucoma, cataracts, and emulsification are inherent in long-term use of silicone oil. Moreover, there is concern that the low molecular weight components lead to ocular toxicity. ²³ Perfluorocarbons, which are relatively well tolerated by the eye, are generally restricted to intraoperative use due to their tendency to emulsify by 2 to 3 weeks. ^{24 25} Thus, it is obvious that, at the very least, successful long-term vitreous substitutes must be nondegradable and nontoxic. We recognize that the ideal vitreous substitute should further be biocompatible, optically clear, nonsoluble, viscoelastic, thixotropic, and injectable. In addition, the hydrogel should be hydrophilic enough to swell slightly to exert a swelling pressure to tamponade the retina without compromising the vasculature. 

Poly(acrylamide) gels were first reported as possible vitreous substitutes by Muller-Jensen and Kohler in 1968. ¹¹ We have modified our previously developed poly(acrylamide) hydrogels ^{14 15 16} to contain 4.5% disulfide cross-linker and 3.0% hydrophobic monomer. The hydrophobic monomer was added to introduce sticky ends on the polymer chains that would enable reversible shear thinning due to intramolecular cross-linking. This resulted in a hydrophilic, nonsoluble, injectable polymer that has viscoelastic properties similar to those of native porcine vitreous. It is imperative that any developed vitreous substitute have viscoelastic moduli close to those of the natural vitreous to act as a viscoelastic damper of eye movements, and exert a small excess of osmotic pressure to serve as a tamponade for a detached retina. In other words, the vitreous substitute must be able to absorb forces exerted on the eye and dissipate them as a function of time. This property becomes particularly important in trauma, as it is the viscoelastic nature of the vitreous that may help protect the eye from injury. ² The consistency is a concern, because the substitute is most practically delivered through a small-gauge needle. Oscillatory shear experiments have demonstrated that, with the significant shear stress applied to a cross-linked polymer during injection, it undergoes irreversible changes in viscoelasticity and consequently there is a decrease in shear moduli, with the gel behaving more like a liquid. ⁹  

We have uniquely managed to maintain the desired viscoelastic properties in the face of injection through a small-gauge needle by the essence of our novel hydrogel design. We used acrylamide as our proof of concept as it is readily available and its monomer is a known neurotoxin, and therefore represents the worst-case scenario for toxicity in vivo. This synthesis method could be readily applied to other more biocompatible polymers. We used a disulfide cross-linker to create initial disulfide bridges. These were then reduced by breaking the cross-links to create thiol groups on the polymer chains. The hydrogel was then injected into the vitreous cavity in this soluble form where it underwent oxidation by air; thus, gelation occured in situ with the formation of covalent disulfide cross-links. 

We evaluated different concentrations of prepared copolymeric hydrogels and found the 2% vitreous substitute sample to best approximate the storage and loss moduli of the natural vitreous humor. The average molecular mass of the polymer was 207 kDa with 0% below a molecular mass of 10 kDa. Given that the acrylamide monomer is toxic and carcinogenic in itself, we devised a rigorous purification process to remove any unreacted monomers. The hydrogel was washed in water numerous times, then reduced to its liquid form, and subsequently dialyzed in water to remove low–molecular weight polymers. The reduced polymer was then precipitated, freeze-dried, and suspended in DPBS. The polymer underwent regelation and formed a remarkably clear gel with a refractive index of 1.338, which is close to that of human vitreous. The uncrosslinked form of the compound could be stored in the presence of nitrogen in a dry form. It underwent autoclaving without any changes in the properties or gel function. In vitro testing demonstrated that the 2% hydrogel formulation underwent gelation within 1 hour under physiological conditions. We found that changing the concentration of the monomer as well as the disulfide cross-linker allowed us to modify this gelation time as well as the viscoelastic properties. 

The hydrogel formulation was easily injected through a 25-gauge conventional needle. Because of its hydrophilic nature, we were able to fill the total vitreous cavity, and a meniscus was not seen on fundus examination. This characteristic may make it superior to current vitreous substitutes for treatment of inferior retinal detachments. Clinical examinations by slit lamp and dilated fundus examination did not reveal any significant anterior segment inflammation, retinitis, vitritis, or gel opacities. The vitreous cavity remained clear in all rabbit eyes up to the maximum follow-up of 1 week. Cytotoxic effects would be expected to be observed initially due to diffusion of low-molecular-weight components through the ocular tissue. One rabbit had an iatrogenic retinal hole with a localized detachment that did not progress. This rabbit was excluded from comparison of ERG amplitudes. None of the eyes appeared to exhibit hypotony or hypertony by digital palpation. Although there were mild absolute differences in the flicker and a-wave responses of the vitreous substitute group, these were not statistically significant. Since we found slight similar depressions in the amplitude responses of the air endotamponade subgroup, we postulate that this may be somehow related to the vitrectomy and evacuation of the native vitreous itself. 

Histologic examination by light microscopy of the eyes that received the 2% hydrogel vitreous substitute further confirmed that there was no evidence of cytotoxicity or architectural disorganization of the neurosensory retina. The observed preretinal inflammatory cells seen in a few samples were clustered mainly around the vitrectomy sites, suggesting a localized response to trauma and adjacent spillover into the posterior segment. We acknowledge that the lack of transmission electron microscopy studies could have resulted in our missing subtle evidence of toxicity. However, in this pilot study, it has been shown that the 2% hydrogel evaluated as a vitreous substitute lacks gross toxicity due to the removal of the monomers and low-molecular-weight polymers from the formulation. Toxicity will be additionally evaluated by transmission electron microscopy in future studies. 

The developed copolymeric hydrogel shows great promise as a long-term vitreous substitute. Its viscoelastic nature, hydrophilicity, and high interfacial tension make it an attractive injectable hydrogel. We want to perform further studies with a more biocompatible monomer by the same synthetic route to evaluate its long-term biocompatibility, biodegradability, and ease of removal. Nonetheless, the developed hydrogel as our proof of concept may represent a novel therapeutic for complex retinal detachments, proliferative vitreoretinopathy, proliferative diabetic retinopathy, and macular hole treatment. 

Conclusions

This polymeric hydrogel mimics the optomechanical properties of the native vitreous. The unique ability to inject it in liquid form and undergo gelation in vivo makes it clinically suitable as a permanent vitreous substitute or drug delivery scaffold. Our preliminary findings indicate overall biocompatibility, clinical suitability, and lack of toxicity, demonstrating the developed hydrogel’s potential as a permanent vitreous substitute. Although our rabbit model data do not show an overt toxic effect by clinical examination, ERG, and histology, further in vivo studies on long-term biocompatibility and degradation are needed. Ultimately, the effectiveness and clinical suitability of this hydrogel will be evaluated in the context of vitreoretinal disease. 

 Supported by a merit review grant from the Department of Veterans Affairs (NR); by awards to the Washington University Department of Ophthalmology and Visual Sciences from Research to Prevent Blindness, Inc.; and by National Eye Institute Core Grant P30 EY02687.

 Submitted for publication September 16, 2008; revised November 19, 2008, and January 5, 2009; accepted July 9, 2009.

 Disclosure: K.E. Swindle-Reilly, None; M. Shah, None; P.D. Hamilton, None; T.A. Eskin, None; S. Kaushal, None; N. Ravi, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Acrylamide copolymer regelation procedure. (A) Initial polymerization and cross-linking, (B) reduction, and (C) regelation in vivo. Reprinted with permission from Swindle KE, Ravi N. Recent advances in polymeric vitreous substitutes. Expert Rev Ophthalmol. 2007;2(2):255–265. © Future Drugs Ltd.

Table 1.

 

Animal Study Summary

Sample	ERG (Day 7)	Day 1 Harvest	Day 7 Harvest	Total Harvest
2% Hydrogel	3	5	4	9
3% Hydrogel	1	0	1	1
Air	2	2	2	4
Total	6	7	7	14

Figure 2.

 

Viscoelastic properties of the vitreous substitutes compared to the porcine vitreous humor.

Figure 3.

 

External slit lamp examination of an eye with the vitreous substitute on postoperative day 1. The vacuole was an air bubble, but there was no lenticular opacity, the cornea was clear, and there was no evidence of clinically significant anterior chamber inflammation.

Figure 4.

 

Fundus photograph of the macula and periphery of a rabbit eye with the vitreous substitute on postoperative day 1. The vitreous cavity remained optically clear and there was no evidence of vitritis, retinitis, vasculitis, or retinal hemorrhage.

Figure 5.

 

Fundus photographs of macula and periphery in rabbit eyes with vitreous substitute on postoperative day 7. The optical media were clear and there was no evidence of retinitis, vasculitis, or pigmentary changes.

Figure 6.

 

ERG response in hydrogel vitreous substitute group (right eye [OD]) in comparison to control (left eye [OS]).

Figure 7.

 

ERG response in air tamponade group (right eye [OD]) in comparison to control eyes (left eye [OS]).

Figure 8.

 

ERG amplitude response ratio (OD/OS) of the hydrogel vitreous substitute (VS) versus control group.

Figure 9.

 

Light photomicrograph of retina from the hydrogel vitreous substitute eye (top) and respective control eye (bottom).