ocular surface & dry eye
Refining surgical techniques for corneal reconstruction

A growing field in the corneal subspecialty, advances in ocular surface reconstruction are also finding their ways into other subspecialties

Source: Edward J. Holland, M.D.

Numerous disease states and indications may result in a patient’s need for ocular surface reconstruction, among them corneal epithelial defects, corneal thinning/perforation, bullous keratopathy, corneal scarring, Stevens-Johnson syndrome, industrial accidents, or severe dry eye.
In the past, these patients were often relegated to tarsorrhaphy, artificial tears, or penetrating keratoplasty (PK) for symptom alleviation. To be more precise, PK, which may be considered as a final option, can be contraindicated as a standing alone surgery in some of these patients.
Today, treatment options include stem cell transplantation for both unilateral or bilateral disease, said Edward J. Holland, M.D., professor of ophthalmology, University of Cincinnati, and director, Cornea Service at the Cincinnati Eye Institute, Cincinnati. His group is working on a combined procedure using cadaver and living donor tissue, dubbed the Cincinnati Procedure. “These are the most challenging patients we face in cornea. They have the highest risk of surgical and treatment failure,” Dr. Holland said.
When patients do present, especially with with cicatrizing conjunctivitis such as Stevens-Johnson syndrome or ocular cicatricial pemphigoid, they have “severely dry eyes, with poor tear film,” said Esen K. Akpek, M.D., associate professor of ophthalmology, and director, Ocular Surface Diseases and Dry Eye Clinic, Wilmer Eye Institute, Johns Hopkins University, Baltimore. “Microbial growth is likely. Because the ocular surface is so dry, it’s easy for the cornea to have an epithelial breakdown. There is also the stagnant and poor tear film which acts as a ‘pond’ on top of that with the potential for bacterial growth.”
Further, in a disease such as Stevens-Johnson, the patient may be on a topical broad-spectrum antibiotic for example if a bandage contact lens is being applied or has been treated with topical steroids, so what remains on the ocular surface is “extremely resistant microorganisms,” or microorganisms other than the usual bacterial flora such as fungi, Dr. Akpek said.

Amniotic membrane transplantation

Amniotic membrane transplantation is an established technique, approved by the Food and Drug Administration in 2001, and Medicare formally recognized the specific surgical procedure around 2004.
“Amniotic membrane transplantation is still relatively new. It has evolved in the last five years in terms of surgical technique, the indications, and the expansion of use beyond transitional ocular surface reconstruction, into glaucoma and oculoplastics,” said Scheffer C.G. Tseng, M.D., Ph.D., director, Ocular Surface Center, Miami; medical director, Ocular Surface Research & Education Foundation, Miami; and director, research and development, Tissue Tech, Miami.
There are two types of membranes available for transplantation—freeze-dried and cryopreserved (Dr. Tseng holds the patent on the latter). The surgical procedure is used to promote tissue healing, reduce inflammation, tissue scarring (approved for cryopreserved, not freeze-dried), and growth of neovascular vessels, Dr. Tseng said. When used as a temporary graft, it can help avoid some of the problems typically associated with wound healing; as a permanent graft, it can help to replace damaged corneal and conjunctival tissue. In its cryopreserved form, it is known as AmnioGraft and Prokera (BioTissue), and in its freeze-dried form as AmbioDry (IOP Inc., Costa Mesa, Calif.; and Innovative Ophthalmic Products, Costa Mesa, Calif.). Oktos received a compliance letter from the FDA in 2005 regarding false claims about its product. Its newest version, Ambio5, is thicker (at 100+ microns) than its predecessors.
Dr. Tseng noted the freeze-dried form of amniotic tissue “cannot claim any of the above biological actions without a premarket clinical trial.”
“Ambio5 combines amniotic and chorionic membranes together,” Dr. Akpek said. She has been using the Ambio5 allograft since its introduction late last year.
“It’s very good for impending or small corneal perforations associated with for example infectious corneal ulcers, It is thick enough for tectonic uses particularly when a donor cornea is not available for a lamellar keratoplasty,” she said. “AmbioDry 5 provides better anatomic support than a multi-layer regular thickness amniotic membrane grafts for corneal support. It is thicker and easier to handle and suture. Older versions are quite thin.” She prefers to use glue to attaché the regular amniotic membrane grafts for example for pterygium surgeries rather than sutures because of the thickness.
Dr. Tseng also prefers to use fibrin glue to make the whole procedure sutureless, he said. “This use of fibrin glue is off-label, but it cuts down on surgical time and postoperative management,” he said.
He believes using a sutureless procedure in conjunction with Prokera can be a huge advantage, especially in cases of chemical burns. “It’s interesting to bring this to acute care settings,” he said. “There’s also a role for amniotic membrane with non-healing ulcers.”
Other uses (both current and potential) include using amniotic membranes as a treatment for managing pain during radiation therapy for uveal melanoma based on Dr. Paul Finger’s report, Dr. Tseng said. Traditionally, those patients receive radiation, “but it’s painful, so amniotic membrane transplantation might work better. Pain is significantly reduced with membranes.”
Use of amniotic membranes as permanent grafts are also being investigated in glaucoma, in cases of shunt exposure and in oculoplastics where muscle exposure or fornix construction might warrant use, Dr. Tseng said.
Dr. Holland prefers to use living tissue to amniotic membranes for diseases with severe limbal cell deficiency. “Patients with severe limbal cell deficiencies need a source of stem cells. They will fail if an amniotic membrane is used without an additional stem cell transplant because the surgeon is not addressing the problems underlying the disease, namely the stem cell deficiency,” he said.
“There’s a role for amniotic membranes in subacute patients where they can help stabilize the cornea. But if a patient is melting, you want some kind of epithelial to grow. The biggest mistake I encounter is surgeons using amniotic membranes as stem cell procedures. It’s just a substrate with no living cells.” Dr. Holland did note that amniotic membrances may be beneficial in managing acute Stevens-Johnson and corneal necrosis.
Dr. Tseng said newer research (Akari, Ophthalmology, 2009 online ahead of print) indicates earlier intervention might now be possible for people with Stevens-Johnson.
“Transitionally, we spend so much time in reconstruction, we didn’t think about prevention,” he said. Akari et al. used steroid pulse therapy at disease onset “ideally within about one week after drug exposure or 4 days after ocular involvement,” Dr. Tseng said.
Using amniotic membrane transplantation also “allowed patients to regain 20/20 vision without ocular complications. These patients don’t have a need for stem cell transplants anymore. We can treat this disease much earlier in its course and correct vision,” Dr. Tseng said, citing Darren Gregory, M.D.1
“We should all be thinking differently and bringing public and general awareness to these breakthroughs,” Dr. Tseng said. “Maybe this is the way to go forward with managing Stevens-Johnson. This is the first time we can do something in the acute phase of a disease and not have to deal with limbal stem cell deficiencies later.”

Limbal/conjunctival stem cell therapies

Sever ocular surface disease is one of the most challenging conditions that clinicians face. It is usually characterized by persistent epithelial defects, ulceration, conjunctivalization, and chronic inflammation.
“First, with stem cell transplants, cadaver or living, is a lot of the time it’s not the surgical technique that’s causing a failure, but managing the ocular surface as whole that’s missing. People fail to recognize the failure of the entire ocular surface defense,” Dr. Tseng said.
When total corneal failure necessitates stem cell therapies, Dr. Tseng prefers to use cadaver tissue for allografts. “We don’t find it necessary to us living donor tissue in our cases; the key to successful surgery isn’t the technique, but the whole management of the eye. If that’s not taken care of, the surgery won’t work, no matter how talented a surgeon you are,” he said.
He said stem cell transplantation is likely to become a subspecialty within cornea, as “the numbers aren’t there to justify everyone perfecting the technique.”
Dr. Holland works with both cadaver and living donor tissues for his allografts. “The advantage with cadaver keratolimbal allografts is that you always have donor available from the eye bank,” he said. “Tissues are harvested a little differently, as the eye bank doesn’t remove the conjunctiva from the limbus. Ideally, we want donors under 60 years old with shorter storage times (ideally under 5 days). For most keratoplasty procedures, the epithelial is the only thing that’s important for limbal cells.”
He added tissue typing is mandatory to reduce the rejection rate when using cadaver tissue.
For more severe injuries or diseases, such as Stevens-Johnson syndrome “it’s usually the severe conjunctival deficiency that leads to failure of the reconstructive surgery,” Dr. Holland said. In those cases, Dr. Holland’s “Cincinnati Procedure” combines living donor and cadaver tissue, where the living donor tissue (related or self, if possible when there’s unilateral disease) is used for the 6 and 12 o’clock meridians and cadaver tissue at 3 and 9 o’clock. The combination of cadaveric keratolimbal tissue and living donor allografts provides a more stable and stronger environment than either alone, he said.
Dr. Holland said that patients receiving allografts require systemic immunosuppression due to the risk of rejection of this vascularized tissue. We have found that immunosuppression in ocular surface transplantation has minimal side effects, “because our patient population is typically younger [than an organ transplant patient], and does not have multiple organ diseases; there’s no hypertension or diabetes, etc. So in those cases, immunosuppression is safer for our patients.”
Our immunosuppression protocol typically uses a triple therapy combination of a corticosteroid, Prograf (tacrolimus, Astellas Pharma, Deerfield, Ill.) and Cellcept (mycophenolate mofetil, Roche Pharmaceuticals, Nutley, N.J.), pre-treating patients the week before surgery with immunosuppressants. They use anti-inflammatory drugs minimally, and taper the first month post-op, with continued use for up to two years in patients without rejection and chronically in low dose for patients with persistent postoperative inflammation, Dr. Holland said. “We’ve had very good results with this approach,” he said.

Concerns

So-called “regular” corneal transplants are ineffective in cases of ocular surface disorders and severe dry eye, Dr. Akpek said. First and foremost physicians have to consider the preoperative condition of the eye, she said. “If the patient has an ongoing inflammation of the corneal and conjunctival surfaces with a poor tear film such as what happens in the case of cicatrizing conjunctivitis anything you put on the eye is going to fail. If the ongoing or underlying condition is not addressed, and there remains active inflammation, the surgery will fail invariably,” she said. Stevens-Johnson is the worst potential pre-op diagnosis in this scenario, she added.
“The eye is more inflamed, patients tend to be younger, the ocular surface is much drier and all three layers of tear film are affected,” she said. When using cadaveric donor stem cells , and even sometimes when using living related donor tissue, “the match will not be 100%, so almost all patients will require immunosuppressive medicines during the post-operative course. How long can you keep them immunosuppressed without causing trouble?” she said. Younger patients are likely to reject, “just like in kidney transplants” because their immunological system is more active than, say, someone in their 70s or later.
Dr. Holland says inherited congenital aniridia usually appears in the first or second decade of life, with severe ocular surface disease. “These patients have no success with standard keratoplasty because the ocular surface will fail due to the underlying limbal disease,” he said. He also treats a “significant” number of patients with a chronic, but non-specific, dry eye syndrome.
In rare cases of dry eye and Meibomian gland disease the chronic inflammation of the ocular surface can result in limbal stem cell deficiency.
Dr. Akpek prefers to perform a Boston type I keratoprosthesis in patients with aniridia related corneal opacities. She has obtained very good results with keratoprosthesis even in aphakic patients.2

Outcomes

The literature notes success rates from transplanted limbal stem cells can range from 25% to 70%, which Dr. Holland said is not surprising.
“The preoperative diagnosis is going to determine the outcome,” he said. “For instance, patients with severe conjunctival deficiency who undergo stem cell transplantation are going to have a poorer prognosis than those with congenital aniridia. It’s the state of the conjunctiva that determines the outcome.”
Once surgery ends, post-op management for allografts should still focus on the ocular surface, Dr. Tseng said. “We need to be vigilant. We need to pay more attention to those surfaces than we might normally,” he said. “We will be handholding these patients for the rest of their lives. It really requires us as physicians to be committed in the long run. We can’t just do a surgery and think we’re done after a couple of follow-up visits.”
Dry eye disease management is underreported, and under-treated, Dr. Tseng added. “People have a simplistic view that dry eye is usually reserved for menopausal women or post-LASIK. If the eye is dry or tear film not there, anything you do to the cornea will fail. It will have gradually declining results,” he said. “Anything that threatens the cornea is detrimental to its future.
Future advances
Dr. Akpek said the future of limbal and conjunctival stem cell transplantation will come from collecting a patient’s own stem cells and growing them in certain culture media.
“Let’s say you have a child in a burn unit who presents with Stevens-Johnson,” she said. “Before those stem cells are dead, we can remove some of the eye surface material, put it in certain tissue cultures, and be able to transplant the result back in that same patient’s eye once the acute inflammation has subsided.” She equates the technique with women who have their eggs harvested before undergoing chemotherapy, and then have the eggs fertilized in vitro and re-implanted in situ once the chemo course has been completed.
She also thinks the technique has great potential in graft v. host disease. For instance, patients undergo full body irradiation before bone marrow transplant surgery, and chemotherapy to ablate the faulty bone marrow. Then they are transplanted with either a relative’s or an unrelated person’s donated bone marrow. By the time the patient undergoes the transplantation, the eyes are “incredibly dry from the conditioning treatment,” she said. But if the ocular tissue is taken prior to the conditioning, once the patient has recovered from the bone marrow transplant, their own tissue can be used to reconstruct the corneal surface.
Dr. Akpek is working on a joint protocol to test that theory with the bone marrow transplant unit at Johns Hopkins.
Dr. Holland said it’s common to use cadaver/living related tissue in the U.S., but in Europe and Japan ex vivo expansion of stem cells is under investigation. “Some surgeons are using cadaver limbus to grow cells or taking a very small sample of living related donor tissue to grow the cells. In ex vivo expansion, “you grow large sheets of cells with very small amounts of original tissue,” he said.
Experimentation is ongoing with obtaining cells from the patient’s own oral mucosa, and early trials are showing some promise for the autograft procedures, he said.
“I think where we’re going in the future is the autograft procedure. Ideally, we would take the patient’s own bone marrow as a source of stem cells to develop a new ocular surface,” he said.
Alternatively a keratoprosthesis may be the most viable option to restore vision in certain cases of severe ocular cicatricial pemphigoid, chemical burns, or even Stevens-Johnson.
“The newest twist is how we incorporate keratoprosthesis into all of this,” Dr. Holland said.
“When a patient presents with total limbal cell deficiency, how do we decide between keratoprosthesis and transplantation?” he said. If the patient is young and healthy, transplantation can be the better option, as the patient will be off immunosuppressants in 2-3 years and can be stable.
“There’s a lot less follow-up than patients who have a keratoprosthesis,” Dr. Holland said. “However the older the patient is, the more likely comorbidities exist and the more potential problems with systemic immunosuppression. In some cases we combine the techniques of ocular surface transplantation and keratoprosthesis. For instance, some patients with Stevens-Johnson syndrome have severe symblepharon formation and are not candidates for a primary keratoprosthesis. Successful ocular surface stem cell transplantation allows the option of subsequent keratoplasty or keratoprosthesis for visual rehabilitation.”

Editor’s note: Drs. Holland and Akpek have no financial interests related to their comments. Dr. Tseng has financial interests with TissueTech, Inc. and owns the patent on cryopreservation of amniotic membrane. He also receives grants from the National Institutes of Health for his studies.

Contact information

Akpek: 410-955-5494, esakpek@jhmi.edu
Holland: 859-331-9000, eholland@fuse.net
Tseng: 305-274-1299, stseng@ocularsurface.com

References:

[1] Gregory DG. The ophthalmic management of acute Stevens-Johnson syndrome. Ocul Surf. 2008;6:87-95.
[2] Akpek EK, Harissi-Dagher M, Petrarca R, et al. Outcomes of Boston keratoprosthesis in aniridia: a retrospective multicenter study. Am J Ophthalmol 2007;144:227-31.