Christina R. Prescott and James Chodosh

Background

Pellier de Quengsy first proposed the idea of a synthetic or partly synthetic cornea in 1789.¹ Dr. Claes Dohlman began development of the Boston keratoprosthesis in 1965; since then, over 7000 have been implanted.² The Boston keratoprosthesis, when constructed, includes four parts (Fig. 50.1). As configured for implantation, the most anterior component is the optic, which is lathed from a single piece of polymethyl methacrylate (PMMA) to form the front plate and its stem. A corneal carrier button (allograft or autograft) permits suturing of the device to host cornea and sits between the front plate and the back plate, which is also made of PMMA. Finally, a C-shaped, titanium locking ring encircles the posterior optic stem and prevents disassembly once the keratoprosthesis is implanted in the eye. Incremental improvements in the device since its inception include a snap-in, rather than screw-in, assembly; the addition of back plate perforations to improve nutrition to the corneal carrier and reduce keratolysis; inclusion of the titanium locking ring (Fig. 50.2); and, currently awaiting approval from the U.S. Food and Drug Administration, a change in back plate material from PMMA to titanium.^3,4 Other advances include the long-term use of antibiotics and soft contact lenses.⁵ The Boston keratoprosthesis type I is likely the most commonly implanted artificial cornea worldwide. The Boston keratoprosthesis type II, a modified version of the type I designed to fit through closed eyelids, is used less often.

Once the decision is made to proceed with implantation (see Chapter 49: Keratoprosthesis Indications), a choice must be made whether to implant a type I or type II design. Underlying pathology, as well as completeness of eyelid closure, quality and quantity of preocular tear film, and depth of conjunctival fornices must be determined. The type II device is only utilized in cases of corneal blindness in which damage to the ocular surface and adnexa is extensive enough to make successful retention of the type I device unlikely.⁶ Candidate patients for a type II keratoprosthesis typically have ocular surface keratinization, severe aqueous tear deficiency, symblephara, and foreshortening of the fornices.⁷

A detailed preoperative discussion of the risks and benefits of keratoprosthesis implantation is extremely important. Patients must understand that they will need continuous treatment, including daily antibiotics, bandage contact lens wear (for the type I keratoprosthesis), and lifelong follow-up care with a qualified corneal specialist. Long-term follow-up is necessary to maintain patient compliance with medications, recognize and treat indolent infection and early keratolysis, and diagnose new-onset or worsening of pre-existing glaucoma, any of which may compromise the visual gains from the initial surgery and lead to loss of the keratoprosthesis and/or the eye. Patients should also be aware that the cosmetic appearance of the eye will change with keratoprosthesis surgery (Fig. 50.3A), quite notably with the keratoprosthesis type II (Fig. 50.3B). To improve cosmesis following type I keratoprosthesis, some patients choose to wear a painted contact lens. After type II surgery, the cosmetic options are limited to the use of tinted glasses, so patients must receive complete preoperative counseling and be willing to accept this limitation prior to undergoing surgery. For best outcomes in patients with underlying inflammatory and autoimmune conditions, such as Stevens–Johnson syndrome/toxic epidermal necrolysis or mucous membrane pemphigoid, attempts should be made to minimize ocular surface inflammation prior to surgery. This may require use of a systemic immunosuppressive agent, such as mycophenolate mofetil. For chronically immunosuppressed patients, collaboration with a rheumatologist, or other similarly qualified physician, is critical to the long-term preservation of the keratoprosthesis.

Prior to surgery, choices must be made about the specific keratoprosthesis to be ordered and what adjunctive procedures will be performed at the time of surgery. In phakic and aphakic patients, keratoprosthesis power is chosen based on the axial length of the eye. If the patient is already pseudophakic, and the surgeon intends to leave the intraocular lens in place, a pseudophakic keratoprosthesis should be used. If the patient is phakic, lens extraction must be performed at the time of keratoprosthesis surgery. Other procedures performed concurrently with keratoprosthesis implantation may include insertion of a plano intraocular lens, insertion of a glaucoma valve, and possible performance of a pars plana vitrectomy or other retina procedure. In phakic patients, the authors typically plan for implantation of a plano, posterior chamber, intraocular lens at the time of lens extraction, so that the keratoprosthesis power will be appropriate even if capsular support is inadequate to support intraocular lens placement.

Special Considerations for the Boston Keratoprosthesis Type I

As with a standard corneal transplants, the choice of anesthesia for a Boston keratoprosthesis type I depends on both the patient’s ocular condition and general health. For most patients, retrobulbar anesthesia is adequate. Patients are prepped and draped in the standard fashion for ophthalmic surgery, using a povidone-iodine 5% preparation in the eye and 10% preparation to the skin. Prior to the first incision, graft size must be chosen based on the corneal pathology and size. The standard keratoprosthesis back plate is 8.5 mm in diameter, although a 7.0-mm back plate is available for pediatric patients and patients with smaller eyes. Because the keratoprosthesis front plate is 5 mm in diameter, the donor graft diameter should never be less than 7 mm, lest the donor corneal rim be insufficient to allow suture fixation. The donor graft diameter is typically trephined to a size 0.5 mm greater than the host trephination diameter, which should be no less than 8 mm when using an 8.5-mm back plate.

Preparation of the Boston Keratoprosthesis Type I

After creation of a sterile field, the presence of all keratoprosthesis components should be verified by visual inspection prior to proceeding with surgery (front plate, back plate, locking ring, and locking pin, the latter used to assemble the device). The donor cornea is then trephined twice: the inner opening is made with a 3.0-mm skin biopsy punch (packaged by the manufacturer with the keratoprosthesis), and the outer trephination is created with a standard corneal donor trephine after diameter selection, as described above. The decision as to which trephination is performed first depends on individual surgeon preference. If eccentric keratoprosthesis placement is indicated, creation of the central corneal trephination should be correspondingly eccentric, thus ensuring proper positioning of the optic at the visual axis.

The keratoprosthesis is then assembled by positioning the optic with the front plate facing down. This can be done on a manufacturer-provided double-sided adhesive, which secures the prosthesis and thus may make assembly easier. The donor cornea’s 3-mm opening is then positioned over the stem, endothelial side up, and the cornea is lowered over the stem until the epithelial side contacts the back of the front plate. The back plate is then positioned over the stem, with the concave side up, and is pushed down the stem onto the donor cornea using the locking pin. It is unclear whether healthy corneal endothelium is necessary for successful keratoprosthesis implantation,⁸ but a small amount of viscoelastic may be placed on the endothelial surface to minimize trauma. The prosthesis components are then locked into place by placement of the titanium locking ring over the stem. An audible snap, appreciated with gentle downward pressure by the locking pin, indicates proper positioning of the locking ring, although proper positioning of the device should always be confirmed by careful inspection, under magnification, prior to implantation. The fully assembled keratoprosthesis is then placed in corneal preservation medium and kept covered in a sterile container until needed.

Boston keratoprosthesis Type I Surgery

Once the keratoprosthesis is constructed, the patient’s cornea is removed using a trephine and corneal scissors, as in traditional penetrating keratoplasty, and is sent for pathologic assessment. The iris is inspected through the corneal opening, and, if corectopia resulting in iris obstruction of the visual axis through the keratoprosthesis optic is present, an iridoplasty is performed. If the patient is phakic, cataract extraction must also be performed. The need for placement of a glaucoma drainage implant at the time of surgery (or later) provides rationale for implantation of a posterior chamber intraocular lens, although surgeon preference dictates whether an intraocular lens is placed or the patient is left aphakic. If the patient is pseudophakic, the intraocular lens is stable, and the surgeon has chosen the pseudophakic-powered keratoprosthesis, the intraocular lens is left in place.⁹

The assembled keratoprosthesis is brought to the operative field, is placed back plate down in the wound and is secured with 12 interrupted 9-0 nylon sutures or 16 interrupted 10-0 nylon sutures. It is particularly important to place and retain a 2–3-mm corneal shield over the keratoprosthesis optic, once the first four cardinal sutures are placed, to avoid phototoxicity to the retina. All corneal sutures are buried in the host tissue, and the wound is checked for leakage.

The authors’ standard intraoperative medications include peribulbar vancomycin, 25 mg in 0.5 mL, ceftazidime 100 mg in 0.5 mL, and triamcinolone 20 mg in 0.5 mL, all given at the conclusion of surgery. These must be injected in a peribulbar fashion to avoid ballooning of the conjunctiva and subsequent difficulty in placing a contact lens. The Boston keratoprosthesis is provided with a 16 mm diameter, 9.8 base curve Kontur™ contact lens (Hercules, CA), which is placed over the eye at the conclusion of surgery and remains in place postoperatively. A semi-pressure patch is then placed, followed by a Fox shield. Ophthalmic ointment is unnecessary and should not be used so as to avoid displacement of the contact lens.

Most patients tolerate the procedure very well and have minimal postoperative pain, similar to the nominal pain experienced after penetrating keratoplasty. On the first postoperative day, topical prednisolone acetate 1%, and a topical fluoroquinolone, both administered four times daily, are initiated. Patients also begin topical vancomycin, 14 mg/mL with benzalkonium chloride preservative, administered once daily within the first week of surgery. The prednisolone and fluoroquinolone are tapered to once-daily administration over 2–3 months following surgery. The authors recommend use of at least one drop per day of a broad-spectrum topical antibiotic long term, and for those cases in which secondary infection risk is of particular concern, additionally one drop of topical vancomycin daily. The long-term use of topical corticosteroid is unnecessary in many cases. More frequent use of topical antibiotics is also unnecessary in the long term and may encourage fungal contamination of the contact lens and/or front plate, with subsequent infection.

Special Considerations for the Boston Keratoprosthesis Type II

The Boston keratoprosthesis type II design is comparable to the type I design, with the exception of an anterior extension of the optic to allow implantation through surgically closed eyelids. The device is prepared similarly to the type I keratoprosthesis, and the corneal component of type II surgery is analogous to type I surgery. However, the remainder of the surgical procedure is much more extensive. General anesthesia is indicated in type II implantation, due to the longer duration of the surgery and the extent of periocular tissue dissection involved.

Boston keratoprosthesis Type II Surgery

Because implantation of the Boston keratoprosthesis type II requires complete closure of the eyelids, avoidance of encystment requires removal of all ocular surface epithelium. Prior to trephination of the cornea, existing symblephara are divided, and conjunctival mucosal epithelium is removed with sharp dissection, including bulbar, foniceal, and tarsal conjunctiva. The upper and lower eyelid margins are infiltrated with 1% lidocaine containing epinephrine, and the eyelid margins are excised, with care taken to fully excise all eyelash follicles. After the host corneal trephination diameter has been chosen, the cornea is marked with the appropriate trephine, and the limbal and corneal epithelium peripheral to the marked area are removed using sharp dissection. The host cornea is trephined and removed as above. If the eye has not previously undergone intraocular surgery, it may be reasonable to minimize additional procedures inside the eye (although, the crystalline lens must always be removed). However, since most patients receiving a type II keratoprosthesis have had prior surgery, an aggressive approach is recommended in order to prevent postoperative glaucoma. These patients might benefit from total iridectomy, removal of the lens and its capsule, pars plana vitrectomy, and posterior placement of an Ahmed valve. Such additional procedures often require involvement of both a vitreoretinal and a glaucoma surgeon. The prepared keratoprosthesis is typically secured with 12 interrupted 9-0 nylon sutures. Knots are rotated posteriorly but need not be buried. However, as in keratoprosthesis type I surgery, it is important to keep a cover over the anterior keratoprosthesis optic throughout the surgery to prevent light toxicity to the retina (Fig. 50.4A).

After implantation of the keratoprosthesis, peribulbar antibiotics and corticosteroid are administered as above, and the eyelids are surgically closed around the optic. Two or three interrupted 6-0 Vicryl sutures are placed on each side of the keratoprosthesis (medial and lateral) using partial-thickness tarsal bites. Once the upper and lower tarsi are brought together on either side of the keratoprosthesis stem, the eyelid margins can be gently closed in mattress style with bolsters and 8-0 nylon sutures. Finally, a notch in the upper lid is fashioned with Vannas scissors (Fig. 50.4B) to allow the keratoprosthesis nub to protrude between the eyelids. It is important that this notch be positioned with the eye in primary gaze so that the keratoprosthesis and eyelid opening are properly aligned.

Prior to reversal of general anesthesia, retroseptal antibiotics and corticosteroid are administered as above, and a retrobulbar anesthetic may be injected to reduce postoperative discomfort. Antibiotic ointment is administered over the skin closure prior to placement of a gentle patch and Fox shield. Postoperatively, the same eye drop regimen described after type I keratoprosthesis placement is utilized. Antibiotic ointment to the eyelid margins is discontinued 2 weeks postoperatively, at which time the skin sutures and bolsters are removed. If present, elevated intraocular pressure is treated with oral acetazolamide or methazolamide. Topical fluoroquinolones and vancomycin are tapered to twice-daily administration after several weeks, and continued indefinitely to reduce microbial contamination of the skin around the keratoprosthesis optic. Once the eyelid skin has fully healed around the keratoprosthesis, usually by 2–3 weeks postoperatively, topical medications do not penetrate to the eye. Topical glaucoma drops are therefore ineffective in reducing intraocular pressure.

Conclusion

Boston keratoprosthesis implantation is a procedure that can be performed by any experienced corneal transplant surgeon, and an important treatment option for patients with severe corneal pathology, who are motivated to improve their vision. Recipient eyes require lifelong care and follow-up.

References