Twelve-year follow-up of penetrating keratoplasty.

PURPOSE: To evaluate the long-term outcomes of penetrating keratoplasty (PKP) according to the corneal disease diagnosis and the number of PKP procedures performed.

METHODS: Five-hundred-and-nine eyes from 403 patients who underwent PKP at Miyata Eye Hospital in Japan from 1998 through 2014, were included in this study. Medical charts were retrospectively examined to ascertain the corneal disease diagnosis and the period of graft survival. Graft survival rates were compared among various corneal disease diagnoses and among the number of PKP procedures performed. Changes in corneal endothelial cell density (ECD) were analyzed using a mixed-effects model. The presence/absence of various risk factors was compared between transparent grafts and failed grafts.

RESULTS: The overall rate of graft survival at 12 years was 60.4%. The rates of graft survival in keratoconus was 100%, in corneal dystrophy 100%, in leukoma 70.8%, and in bullous keratopathy 51.7%. The rates of graft survival at 12 years for the first PKP was 65.4% and for the second PKP, 43.4% (p < 0.001). All cases of third PKP and fourth PKP failed within 8 years. Preoperative mean ECD (95% confidence interval) was 2722 (2666-2778) cells/mm(2); it decreased exponentially after PKP. Mean ECD was 659 (440-878) cells/mm(2) at 10 years. Rejection, trauma, and infection occurred significantly more frequently in failed grafts than in transparent grafts.

CONCLUSIONS: The long-term prognosis of PKP depends on the original diagnosis. The long-term prognosis of re-grafting is worse than that of primary grafts. Rejection, trauma, and infection are risk factors for graft failure.