Special Report

Tools for Challenging Cases

Repairing the ocular surface with amniotic membrane grafts and stem cell transplantation.

By John D. Sheppard, MD, MMSC

Some of our most challenging cases are those that require multiple, staged surgeries and years of follow-up. These cases often involve vision loss or even risk of blindness. When we need to perform a restorative operation and promote healing of the ocular surface, two of our most powerful and exciting tools are amniotic membrane grafting and limbal stem cell transplantation. Both procedures help the ocular surface to regenerate as these challenging and rewarding patients begin the long journey to stable, healthier eyes and improved vision.

Why Use Amniotic Membrane?

The characteristics of amniotic membrane make it ideal for grafting this tissue. The membrane may aid in “scarless wound healing” after fetal surgery, and it may impart those same anti-cicatricial properties to the ocular surface. Amniotic membrane contains an array of polypeptide factors that inhibit necrosis through enzymatic deactivation, while promoting healing with trophic proteins. It is durable, and it is not rejected.

A wide variety of conditions are amenable to treatment with amniotic membrane. Currently, we tend to utilize amnion in our practice for significant pathologic conditions such as high-risk corneal surgery grafting, stem cell deficiency, necrotizing infection and persistent non-healing ulcers.

Cases for Amniotic Membrane

I use amniotic membrane to treat many conditions. For example, any time I encounter a severely necrotizing infection with loss of stromal tissue, I sterilize first with antibiotics or antivirals, then use amniotic membrane to stop stromal keratolysis. If a perforation is threatened or a protruding Descemetocele is apparent, cyanoacrylate tissue adhesive may be necessary. Amniotic membrane procedures can then follow the stop-gap gluing surgery if necessary. We've found that smaller corneal defects below 0.5 mm actually respond well to organic tissue glue such as Tisseel (Baxter) and Evicel (Ethicon), and a very small piece of the patient's own conjunctiva taken from a superotemporal donor site. It is imperative to deflate the anterior chamber, then insert sterile air through a paracentesis in order to keep the Seidel positive leak dry during the tissue glue, amniotic membrane or cyanoacrylate repair procedure.

Amniotic membrane presents numerous unique beneficial characteristics for healing deficit cases as well. We recently treated a patient with an exquisitely painful Mooren's ulcer. At 2 weeks, she was melting essentially all of her corneal tissue, accompanied by extreme pain and marked diffuse anterior scleritis. She was not responding to oral doxycycline and prednisone and numerous antimicrobial topical medications, including azithromycin. I performed amniotic membrane transplantation, and for the first time in 6 weeks, the patient was pain free. She began the healing process necessary for an anterior segment reconstruction in the future.

A 35-year-old African American female with a 4-week history of severely painful culture and biopsy-negative keratitis was unresponsive to antimicrobial therapy. The entire stroma was melting and there was an enlarging descemetocele.

I also use amniotic membrane to help avoid corneal perforation in several ways. For example, one of my patients had gram-negative bacterial keratitis (pseudomonas) with rapid thinning. Steroids promote the activity of gram-negative proteolytic enzymes, so I saturated this eye with antibiotics and performed amniotic membrane transplantation to avoid perforation and control the inflammation with the amnion.

Amniotic membrane is effective for neurotrophic problems, such as herpes zoster or chronic herpes simplex stromal keratitis. I use an amnion graft with tissue glue or sutures, along with antimicrobial therapy, and the eye usually is able to resurface and heal more rapidly.

Amniotic membrane has other indications for less potentially blinding conditions as well. For example, for pterygium and conjunctival chalasis, I first remove excessive redundant or pathologically vascularized conjunctiva. Next, I stretch and reposition the conjunctiva if possible, or I can place a graft of the patient's own conjunctival tissue or amniotic membrane over the defect. The graft can be sewn in place, but because it's often so large, I like to sew anchor corners to ensure that they don't loosen, and use glue for optimal adhesion and healing. Of the two tissue glues, Tisseel and Evicel, the latter is more amenable to ocular surface surgery. I prefer 10-0 nylon, which is easy to manipulate under the microscope, does not instigate inflammation and the knots can usually be buried.

For my patients with conjunctival and corneal ocular surface surgery, including amniotic membrane grafts, I use a therapeutic bandage contact lens to promote healing and diminish the effect of blinking on the ocular surface. This simple final step in the operating room has greatly contributed to patient satisfaction by accelerating epithelialization of the cornea and significantly reducing postoperative pain. We recommend a therapeutic bandage contact lens for a wide variety of ocular surgeries, including corneal transplantation, Descemet's stripping with endothelial keratoplasty, lamellar keratectomy, pterygium, amniotic membrane transplantation, cyanoacrylate glue and complicated cataract surgery where multiple limbal incisions may be required. It is also wise to use a bandage lens to make even routine premium intraocular lens patients as comfortable as possible during their first post-operative evening at home.

Cost and Availability

In our practice, we tend to use amniotic membrane grafts for only the most significant pterygium and conjunctival pathology, particularly because of the cost of amniotic membrane and organic fibrin tissue glue incurred by non-hospital surgical facilities. In the ambulatory surgery setting, reimbursement for the procedure is 60% less than if the identical operation were performed in a hospital. We can't pass along the cost of materials to patients, nor is the amnion considered a “pass-through” item. Therefore, the cost must be absorbed by the ambulatory surgery center within the profit margin of the reduced facility fee. So to avoid losing revenue with amniotic membrane cases in the private setting, we're forced to move these cases to the hospital, which means a significant increase in turnover time.

Two days following subtotal lamellar keratectomy and repair with 12 mm diameter amniotic membrane using 9-0 Vicryl sutures, the adjacent scleritis was subsiding rapidly and tissue necrosis ceased. Her pain had almost completely subsided.

Although there are several forms of readily available dry amniotic membrane, including AmbioDry (IOP Ophthalmics) and AmnioGraft (Bio-Tissue), a likely combination of regulatory barriers and hospital bureaucracies make it impossible to use living donor tissue for the conjunctiva. Other countries have less stringent impediments to gaining convenient access to living, fresh amniotic membrane.

As a visiting professor in Chieti, Italy, I was pleasantly surprised to learn of the advanced method of amniotic membrane procurement used by surgeons there. Ophthalmologists collaborated with the obstetrics department to obtain fresh, potent, amniotic membrane from exquisitely screened candidates, straight from the delivery room. After sterilization with povidone iodine, surgeons performed the graft, and their patients healed reliably and rapidly with this “superhero” amniotic membrane.

One new novel form of amniotic delivery, ProKera (Bio-Tissue), is a dome of amniotic membrane supported by a ring contact lens. We can place this human tissue contact lens on the eye with arrested healing in the office, bypassing a procedure in the hospital altogether. However, this modality was not funded in the clinic until early 2011. It is also funded in the ASC. A careful lamellar keratectomy and a steroid injection can be performed in an ASC as well.

Limbal Stem Cell Transplantation

The patient population for limbal stem cell surgery on the ocular surface is so heterogeneous and the conditions are so rare that it's actually very difficult to put together a clinical study for FDA approval. The wide variety of applications makes cases different, such as alkaline or acid burns, severe surface injury, stem cell dropout, severe medication toxicity, limbal stem cell deficiency with vision loss, Stevens-Johnson syndrome, ocular cicatricial pemphigoid, or a litany of ocular surface diseases for which there currently is no standard treatment.

These are some of the most challenging, frustrating surgeries we face, all with lengthy follow up, and many patients requiring life-long follow-up visits. Needless to say, there are significant insurance funding challenges with these patients as well.

The best source for limbal stem cells is the contralateral eye. The sample is fresh, there's no risk of rejection, and bureaucratic and financial impediments are eliminated. I like to take a section of limbal conjunctiva and graft it into the damaged eye. Inferior bulbar conjunctiva donor tissue from the redundant cul-de-sac also suffices in most cases and is technically much easier to procure. Obviously, we can't resurface the recipient's entire eye, but in combination with amniotic membrane and a therapeutic soft contact lens, the surgery is frequently successful.

If it's not possible to sample stem cells from the con-tralateral eye, the next most reliable source is a living related donor, preferably a sibling. The beauty of a compatible living stem cell donor is that the patient needs little or no systemic immunosuppression. Otherwise, we have to load up recipients with potent systemic immunosuppressive agents to inhibit tissue rejection very much in the same way we prevent rejection of solid organ transplants, such as livers or kidneys. However, there are barriers to finding related donors, such as geographic or emotional distance from family members or philosophical hesitancy. In addition, there is the considerable expense of tissue typing when unrelated living or cadaveric donors are the only available tissue source. Unlike corneal transplants, conjunctival tissue is highly vascular and therefore not at all resistant against rejection.

To graft donated stem cell tissue, we excise the donated sections 30 minutes before the grafting procedure. Because patients who require donors typically have contralateral damage, we must choose which of the patient's eyes should receive surgery first. Depending on the status of each eye, we may start with either the more damaged eye or the one with a better chance of recovery. Significant ocular surface disease may also be accompanied by advanced anterior segment cicatrisation, synechiae and trabecular meshwork damage, as often seen in severe alkaline burns, for example. Thus, the less severely afflicted eye may be chosen for surgery because the more damaged surface is also accompanied by end-stage glaucomatous optic atrophy. Surgery begins with a resurfacing procedure where we remove non-viable tissue, after which we perform four quadrantic limbal grafts of stem cell tissue, as well as amniotic membrane grafting covering the entire damaged surface area. It's an exciting and very technically challenging staged resurfacing process, often leading to a corneal transplant months later. The follow-up care is equally challenging. Co-management of systemic immunosuppressive agents with a qualified immunologist who is experienced in solid organ transplant rejection management markedly improves the prognosis, enhances patient safety and simplifies care for the operating ophthalmologist.

A Future Without Donors?

When we're unable to secure a related tissue donor and the patient is facing a choice between blindness and independence, we move to unrelated donor tissue. Managing the surgery requires meticulous attention to detail and close collaboration with a team to manage the immunosuppressed patient.

It is fascinating to know that basic science researchers are making rapid advances in producing stem cells for the eye from progenitor ocular cells as well as other parts of the body. Target tissues include the ocular surface, the corneal endothelium and the retinal pigment epithelium. Patients would be able to receive a cultured stem cell graft with the same success as a graft from their own eye. The ideal location for harvesting these stem cells for the ocular surface is the buccal mucosa, blood, bone marrow or even dermal fibroblasts. An advance like this would mean more compatible tissue grafts made easily and practically, without the ethical challenges of harvesting fetal stem cell tissue. In the near future, routine repair of the ocular surface may be possible with cultured organs from advanced stem cell biotechnology.

John D. Sheppard, MD, is President of Virginia Eye Consultants in Norfolk, Va. In addition, Dr. Sheppard is Professor of Ophthalmology, Microbiology & Molecular Biology, Ophthalmology Residency Research Program Director, and Clinical Director of the Thomas R. Lee Center for Ocular Pharmacology at Eastern Virginia Medical School.