In Vivo Rabbit Eyecup Preparation for Use in Retinal Research

Abstract

purpose. To develop an in vivo rabbit eyecup preparation that preserves neuronal and vascular connections with normal posterior segment contour, permitting direct access to the retina and facilitating retinal microsurgical and neuroscience research.

methods. Cyanoacrylate glue was applied to the anterior sclera of six Dutch-belted rabbits before open-sky vitrectomy. The glue was used to harden the compliant scleral wall and to fix it to the surrounding periorbital tissues.

results. A normal contour of the posterior segment was retained in all cases. Vitrectomy under air was successfully accomplished and an extensive removal of the vitreous gel was achieved. Fluorescein angiography revealed normal vascular patency of the retinal vessels after vitrectomy.

conclusions. The proposed modification of the rabbit eyecup retains the normal neurovascular connections and configuration of the retina, making it suitable for retinal microsurgical maneuvers or any procedures in which easy access to anatomically undisturbed retina is required.

The development of new retinal surgical techniques, as well as accurate recordings from retinal neurons, can often be facilitated through the use of an open-sky approach. Visualization is improved through the elimination of anterior segment structures, and simplifying surgical manipulation by using a large opening rather than through small sclerotomies allows the researcher to concentrate on retinal tissue characteristics rather than on the mechanics involved in a pars plana approach. In addition, direct illumination of an open-globe preparation with an operating microscope light allows for easier bimanual surgical maneuvers. Although human cadaveric or enucleated animal eyes are useful in many of the early stages of development, research in new surgical techniques often requires an in vivo model. In addition to decreasing artifacts associated with nonviable retinal tissue, an in vivo model such as a rabbit eyecup, allows for measurement of electrophysiologic and ionic conditions, as well as the effect of active perfusion of the tissue. ^{1 2} Rabbit eyecups have been described in the past, in which fixation rings are sutured onto the episclera after removal of anterior segment structures. ³ Because of the decreased rigidity of a rabbit scleral wall compared with that of a human, the posterior segment is often distorted from scleral folds, making retinal manipulation more difficult. Unfortunately, this characteristic can severely limit the usefulness of such an open-sky model, depending on the surgical technique being investigated. We describe herein a modification of the eyecup preparation in the living rabbit that greatly decreases the amount of posterior segment distortion. It has the additional benefits of ease of setup, and still retains all the advantages of an in vivo model. 

Methods

Six Dutch-belted rabbits (3–5 lb) were used, in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Each rabbit was anesthetized with an intramuscular injection of ketamine hydrochloride (50 mg/kg) and xylazine (20 mg/kg) and placed under an operating microscope (M651; Wild Leitz USA, Inc., Rockleigh, NJ). Both the upper and lower lids were excised, and a 360° conjunctival peritomy performed using Westcott scissors. The sclera was then exposed in each quadrant, and a small amount of cyanoacrylate glue (CVS Super Glue Gel; CVS, Woonsocket, RI) was gently spread over the scleral surface anterior to the equator. After the glue was allowed to harden for 2 to 3 minutes, the superior and inferior rectus muscles were grasped with forceps, and gentle anterior traction was applied to the globe. Simultaneously, additional glue was applied around the globe up to the orbital rim. Anterior traction on the globe was maintained until the glue had hardened. It is important to note that insufficient or excessive traction both result in the formation of folds. Subsequently, the anterior chamber was entered using a no. 11 surgical blade, and corneoscleral scissors were used to extend the wound from this site to create a circular incision 1 mm anterior to the limbus. This resulted in the removal of the entire cornea (Fig. 1) . Four equally spaced radial incisions were made in the iris to the iris base, and the resultant iris flaps reflected back (Fig. 2) . Additional cyanoacrylate glue was placed evenly over the reflected iris to quickly achieve complete hemostasis in any severed iris vessels. Surprisingly, very little bleeding occurred, even before application of the glue. In total, approximately 1 g cyanoacrylate was used for each eyecup preparation. The anterior capsule of the lens was opened with fine forceps, and the lens was expressed in an extracapsular fashion. Vitrectomy was performed (Storz Premier Microvit; Storz Instrument Company, St. Louis, MO) with a cut rate of 600 cuts per minute (cpm) and an aspiration rate of 150 mm Hg. The vitrectomy was performed under air without fluid infusion (Fig. 3) . Residual fluid produced by the ciliary body was aspirated with a soft-tipped cannula. 

In two animals, fluorescein angiography was performed at the end of the procedure with a fundus camera (TRC 50 VT; Topcon, Paramus, NJ), mounted with an adapter containing an achromatic lens with a focal length of 25.4 m (PACO 22; Newport, Irvine, CA). The image was digitally captured and analyzed on computer (Imagenet for Windows, ver. 1.53; Topcon). 

At the end of the procedure the animals were killed with an intracardiac injection of 3 to 4 mL pentobarbital sodium (Fatal Plus; Vortech Pharmaceuticals, Dearborn, MI). 

Results

In all cases, the normal contour of the posterior segment was retained, and the retinal surface and vessels remained in a normal anatomic position (Fig. 4) . The retention of a smooth posterior surface, coupled with the excellent visualization of the vitreous afforded by performing the vitrectomy under air permitted an extensive removal of the vitreous gel. Most of the retinal surface was free of vitreous at the end of the procedure as indicated by gentle aspiration using a silicone-tipped cannula. A residual vitreous stalk, however, remained attached at the center of the myelin wing. Fluorescein angiography at the end of the procedure demonstrated normal patency of the retinal and choroidal vasculature (Fig. 5) . 

Discussion

The collapse of the thin scleral walls in the rabbit model has been a significant obstacle in the use of an in vivo eyecup for experimental retinal surgery, as well as for electrical or chemical retinal recordings. Other investigators ³ have used metallic rings sutured at the limbus for structural support of the globe. Although the anterior portion of the sclera remains well supported, these rings fail to maintain the normal contour of the retina posteriorly. In previous attempts, the weight of the globe has caused the posterior globe to collapse, creating deep chorioretinal folds that significantly impaired surgical manipulations of the retinal surface, and make this model unsuitable for performing such maneuvers. The model that we describe uses cyanoacrylate glue to “cast” the anterior sclera, hardening it and preventing its collapse. 

In addition, the glue attaches the sclera to adjacent periorbital structures, including the lid remnants and adjacent soft tissues that serve to suspend and support the entire eyecup. This keeps the weight of the globe from resting on posterior structures and thereby prevents distortion of the posterior half of the globe. Although applying glue to the posterior sclera would harden the sclera in this area as well, glue in this part of the eye impairs normal blood flow. By avoiding application of glue to sclera posterior to the equator, this side effect is avoided, and normal vessel patency is maintained in the eyecup. The technique is simple, can be set up quickly without suturing, and permits exposure of the posterior segment with minimal distortion compared with that created by fixation rings from suture traction. 

By avoiding incisions in the sclera, this technique also minimizes bleeding from episcleral vessels. The minimal bleeding resulting from the iris incisions is quickly controlled by reflecting the iris tissue and applying additional glue. This can be performed very easily without requirement for any additional instruments such as a cauterizer. The removal of the anterior segment optical elements allows for excellent high-magnification visualization under the operating microscope without the need for a vitrectomy contact lens. 

The relatively large lens of the rabbit makes complete removal of the vitreous impossible with a three-port, pars plana approach. Our method obviously does not have this limitation. In addition, vitrectomy under air facilitates excellent visualization of the vitreous gel, which permits an extensive removal of the gel, leaving the retinal surface apparently free of vitreous on microscopic, but not confirmed by histopathologic, examination. The possibility to perform an extensive removal of the vitreous gel may represent an additional advantage of this model, especially for applications focused on manipulations or recordings on the retinal surface. Total removal of the vitreous and posterior vitreous detachment in the rabbit has been found difficult to perform by pars plana vitrectomy. ^{4 5} The extent of vitreous removal achieved with our model seems to be sufficient for surgical maneuvers on the retinal surface. However, the presence of posterior vitreous detachment must be confirmed in future experiments with histopathological techniques. Other investigators have reported the presence of residual collagen fibers on the rabbit retinal surface, even after an apparently total removal of the vitreous gel. ⁶ Without such histopathological confirmation, we cannot be certain what degree of vitreous removal has been achieved. 

Fluorescein angiography was performed on two animals by using an adapter with an additional lens in front of the fundus camera’s objective lens. The same system has been used successfully for iris angiography in rabbits. ⁷ The fluorescein angiography demonstrated normal patency of the retinal and choroidal circulation. This makes the model suitable for experiments focused on retinal and/or choroidal vasculature and circulation. 

A main limitation of this model is the need to kill the animal after the procedure. This restricts its use in short-term experiments. 

In conclusion, we have demonstrated a modification of the eyecup preparation in rabbits that retains the normal anatomic contour of the posterior segment. Possible applications for this model include terminal experiments of the retinal surface requiring direct observation of the retina surface during microsurgical manipulations such as staining, peeling, and laser ablation of its superficial layers. This eyecup is also useful for developing new microsurgical interventions in the retinal vessels, electrical recordings of retinal neurons, and documentation of the direct action of various medications on the retinal tissue and vasculature. 

 Supported by the Vitreoretinal Research Fund (DJD).

 Submitted for publication April 27, 2003; revised July 23, 2003; accepted July 24, 2003.

 Disclosure: M. Tsilimbaris, None; E.S. Lit, None; J.B. Christoforidis, None; D.J. D’Amico, 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.

 

Rabbit eye after glue application and dissection of the cornea. Eyelids have been excised. Glue has been applied on the sclera and around the globe.

Figure 2.

 

Iris flaps created after four equally spaced radial incisions. The flaps are reflected and the anterior lens surface is exposed.

Figure 3.

 

Vitrectomy probe within the vitreous cavity during open-sky vitrectomy. The glue hardens the eyecup walls and prevents collapse. The myelin wing can be seen at the base of the eyecup.

Figure 4.

 

(A) Eyecup at the end of procedure. Glue support allows the eye to retain its normal shape and configuration. The myelin wing and vessels can be seen at the posterior pole. (B) Larger magnification of the posterior pole in another eye.

Figure 5.

 

Fluorescein angiography performed after the end of the procedure. Normal patency of the retinal and choroidal vasculature is demonstrated. (A) Early phase; (B) late phase.