Amniotic membrane technology may be a welcome addition to the DSAEK technique armamentarium, especially for the majority of cases that already have epithelial edema, haze, and scarring.
By Michael P. Ehrenhaus, MD, and Sebastian Guzman, MD; Special to Ophthalmology Times
Descemet’s stripping automated endothelial keratoplasty (DSAEK) is an excellent choice for many patients-whether they have moderate to advanced Fuchs’ dystrophy alone, are in a post-surgical status from weeks to years later, or have endothelial dysfunction from other contributing factors.
DSAEK is often done with the epithelium intact to help with both the surgical view and with patient discomfort. However, as most of us have experienced, many of these cases are done through a loose and often hazy epithelial sheet, either due to long-standing edema, chronic micro or macrobullae or even a coincident basement membrane dystrophy. This may actually end up causing more discomfort though due to increased surgical time as well as possible poorer surgical outcomes.
In addition, an intact epithelial surface may be an important layer that helps with the implanted corneal endothelial discs adhesion. Think of the cornea as a book lying on the ocean, with intact front and back book-covers. The back cover pumps work to keep the books pages clear by pumping out the fluid that is slowly seeping in, while the front cover protects the pages from the elements. In theory, if the front cover is missing, the pump forces have a decreased counter-balancing layer that helps with the back covers pump function to improve the clarity and relative dehydration of the stroma.
However, in cases where the epithelium is cloudy, loose, edematous or scarred, by removing the epithelium your surgical view is dramatically improved. Difficult cases now become more “routine” because you can now better appreciate the anatomy that you need to work with.
The problem with epithelial removal has been greater post-operative discomfort causing patients to squeeze their eyes more often and more forcefully which we know is a contributing factor to endothelial disc dislocation off of the posterior stroma. Additionally, the proposed decreased counter-balancing force utilized by the endothelial pump layer is missing, which is needed in theory to improve the pump gradient and help with adherence, may lead to dislocations as well.
By utilizing the well-known properties associated with preserved amniotic tissue and epithelial healing, as well as covering the corneal epithelial defect with a bandage contact lens, the surface counter-balancing layer can be re-established, the patient’s discomfort can be minimized, and the epithelium can more rapidly heal, allowing for both improved surgical outcomes as well as improved postoperative visual acuities.
To prove this ourselves, we perfomed DSAEK surgery on a more routine post-phaco bullous keratopathy as well as a more complex failed prior DSAEK case that also had other prior complicating surgical history with resultant irido-corneal adhesions. Both cases had increased epithelial clouding and looseness. During the surgery, the epithelium was removed to drastically improve the surgical view which allowed for a complete Descemet’s layer removal, appropriate synechiolysis, and enhanced viewing of the endothelial disc insertion and centration.
At the end of the cases, we placed an AmbioDisk with an Acuvue2 soft contact lens onto the surface of the cornea. The Amniotic disc in these cases was created by using an 8-mm trephine to cut an Ambiodry2 graft. The amniotic disc was placed to the surface of the cornea, covering the epithelial defect and antiotic drops and BSS drops were placed to hydrate it slightly. The contact lens was soaked in topical antiotic drops and then placed to the surface of the eye as well prior to placement of a patch and shield.
Postoperatively, the patients stated that their eyes were comfortable both immediately after surgery as well as during the visits the following day and thereafter. During the first week, the AmbioDisk and contact lens remained intact and well centered. Visual acuity was only slightly decreased initially due to the epithelial defect, the thin, slightly opaque amniotic tissue layer and the bandage contact lens, but the donor tissues were attached 100%, centered and already thinning nicely. When the lens and tissue were removed at 1 week, the epithelium had healed completely and the visions were dramatically improved from both pre-operative baseline and on postoperative day 1.
We plan to do more cases this way and feel it will prove to be a welcome addition to the DSAEK technique armamentarium, especially for the majority of cases that already have epithelial edema, haze, and scarring. We look forward to enjoying the improved surgical view and improved outcomes, without the worry of possibly causing disc locations as a direct result of the epithelial removal.
Michael P. Ehrenhaus, MD, is director, New York Cornea Consultants, Bayside, NY. Readers may contact him via phone at 718/428-8400 or e-mail at firstname.lastname@example.org. Dr. Ehrenhaus is a speaker/consultant for Allergan and Bausch + Lomb.
Sebastian Guzman, MD, is director, ophthalmology, Clinica Corominas, Santiago, Dominican Republic. Readers may contact Dr. Guzman via phone at 809/971-2020 or e-mail at email@example.com. Dr Guzman has no financial interests.
1. Dua HS, Gomes JA, King AJ, Maharajan VS. The amniotic membrane in ophthalmology. Surv Ophthalmol. 2004;49:51-77.
2. Tseng SC, Espana EM, Kawakita T, Di Pascuale MA, Li W, He H, Liu TS, Cho TH, Gao YY, Yeh LK, Liu CY. How does amniotic membrane work? Ocul Surf. 2004;2:177-187.