Outcome of graft central thickness after penetrating keratoplasty

Objective: To analyze the outcome of graft central thickness after penetrating keratoplasty. Design: Retrospective observational cohort study. Participants: Eight hundred fifty-six consecutive penetrating keratoplasties (772 patients) performed between 1992 and 2001 were analyzed and 772 were included (1 graft per patient was included). Methods: Slit-lamp examination and ultrasound pachymetry. Main Outcome Measures: Slit-lamp findings, intraocular pressure, and graft central thickness were recorded. Results: Patients were observed for an average of 55 months. The 36- and 60-month graft survival estimates were, respectively, 77.2% and 71.0%. The average graft central thickness in successful transplants was 655 mu m at 1 week, 558 mu M at 1 month, 533 pm at 6 months, 538 mu m at 12 months, 558 mu m at 24 months, 561 mu m at 36 months, and 568 mu m at 5 years. At each postoperative time point, the percentage of eyes with decreased, normal, and increased graft central thickness was significantly different according to slit-lamp findings. Of the patients with increased graft thickness, 46.2% had a simple outcome (normal intraocular pressure and normal slit-lamp findings), 28.6% experienced rejection, 15.8% experienced a graft nonimmunological event, and 9.4% experienced increased intraocular pressure. At each postoperative follow-up, subsequent graft survival was significantly lower in patients with increased graft thickness as compared with patients with normal or decreased graft thickness. When analyzing only patients with simple outcome, the relative risk of graft failure was 3.3 if graft thickness was increased at 1 month (P < 0.0001). Conclusions: In conclusion, graft central thickness assessed by ultrasound pachymetry is a useful method for observing patients who have undergone penetrating keratoplasty. Even when slit-lamp examination reveals no complications, patients with an increase in graft thickness above the upper limit of normal for the postoperative time point under consideration are at greater risk of failure. Ophthalmology 2005;112:626-633 (c) 2005 by the American Academy of Ophthalmology.