Penetrating keratoplasty for endothelial decompensation in eyes with buphthalmos

To evaluate the prognosis and complications of penetrating keratoplasty (PKP) for corneal decompensation in eyes with buphthalmos and to analyze the risk factors for graft failure.

Clinical records of 13 adult and three pediatric patients who underwent PKP for endothelial decompensation with a previous diagnosis of congenital glaucoma of a total of 3,663 corneal transplantations performed in our department between January 1987 and December 2001 were reviewed retrospectively. During the study period, a total of 33 PKPs was performed in 20 eyes with buphthalmos. The median age of the patients at the time of PKP was 39 years (range, 3 to 72). All patients had a history of intraocular surgery, including multiple glaucoma surgeries, cataract extraction, and PKP. The impact of pre-, intra-, and postoperative factors on graft failure and duration of graft clarity was analyzed.

Fifty-five percent (11/20) of the eyes received only one graft, 25% (5/20) received two, and 20% (4/20) received three grafts. During a mean follow-up of 87.2 months (range, 4.5-72), graft failure occurred in 18 of 33 grafts (54%). Seven (7/18, 39%) had immunologic graft rejection, and 11 (11/18, 61%) had nonimmunologic graft failure. At the end of the follow-up, 75% (15/20) of the eyes had clear grafts. Duration of graft clarity was found to be significantly shorter in regrafts compared with that of primary grafts (27.0 +/- 27.7 versus 56.4 +/- 41.0 months, p= 0.02). After PKP, intraocular pressure (IOP) was uncontrolled in 12 (12/33, 36%) grafts. Nine of 20 eyes (45%) required an average of 3.2 cyclodestructive procedures per eye for pharmacologically resistant elevated IOP. The final postoperative vision improved in 70% (14/20) of the eyes and the best visual acuity postoperatively (75% > or =20/400) was significantly better than the preoperative visual acuity (25% > or =20/400, p= 0.0001).

Endothelial decompensation due to congenital glaucoma is a very rare indication for PKP. The incidence of graft failure is high, and nonimmunologic reasons are the leading causes of graft failure in this high-risk population. Visual acuity can be significantly improved but is usually still very limited by advanced glaucomatous optic nerve damage and amblyopia. Efficient control of IOP before and after PKP is mandatory in eyes with buphthalmos to avoid graft failure and progress of glaucomatous optic nerve atrophy.