The cornea is a clear, dome-shaped membrane covering the front of the eye; its layers include the epithelium, Bowman layer, the stroma (≈90% of corneal thickness), Descemet membrane, and the endothelium. The endothelium maintains corneal clarity by removing and limiting fluid in the stroma; endothelial dysfunction (for example, Fuchs endothelial dystrophy, aphakic or pseudophakic bullous keratopathy, or failure/rejection of a prior graft) leads to corneal edema and loss of transparency.
Traditional penetrating keratoplasty (PK) replaces full-thickness donor cornea and may require prolonged visual rehabilitation and carries risks such as irregular astigmatism from sutures, wound dehiscence, endophthalmitis, expulsive suprachoroidal hemorrhage, and catastrophic wound failure. Posterior lamellar approaches—collectively called endothelial keratoplasty (EK)—selectively replace the diseased endothelium and Descemet membrane. Techniques include DSEK, DSAEK, DMEK, and DMAEK, with donor tissue prepared by hand, microkeratome, or laser. EK is associated with faster visual recovery (typically 4–8 weeks), less postoperative astigmatism, and avoidance of full‑thickness wound complications, but has procedure-specific risks such as graft dislocation, endothelial cell loss, increased intraocular pressure, graft rejection, and late endothelial failure and may require repositioning in the early postoperative period.
Endothelial keratoplasty is considered standard for many endothelial disorders; however, laser-assisted variations (femtosecond or femtosecond/excimer assisted EK) are considered investigational in this policy. As a surgical procedure, EK itself is not FDA-regulated, although microkeratomes used in donor preparation have received FDA 510(k) clearance.
Keratoprostheses are artificial corneas intended for patients with severe corneal opacity who are poor candidates for standard transplant. The Boston (Dohlman‑Doane) Keratoprosthesis (Boston KPro) has FDA premarket approval (1992) for use when standard corneal transplant has failed or is unlikely to succeed; case series and systematic reviews show improved visual outcomes but high complication rates and the need for careful patient selection and experienced centers. The AlphaCor device was cleared via 510(k) (August 2002) and has been associated with limited evidence of benefit and notable complications (anterior surface thinning/melting and necrosis).