Long-Term Corneal Endothelial Cell Counts After Penetrating Keratoplasty in Infants

Purpose To report long-term endothelial cell counts after penetrating keratoplasty (PKP) in infants. Methods The charts of all children who have undergone PKP in their first year of life between 1998 and 2013 at the Hospital for Sick Children, Toronto, Canada, were reviewed retrospectively. Patients who had a single successful transplant with a valid endothelial cell density (ECD) analysis postsurgery were included in the study. Donor ECDs were provided by a local eye bank. Specular microscopy images were taken at a variable interval after surgery using a noncontact specular microscope (ROBO, Konan; Konan Medical) and endothelial cell loss was calculated. In young children images were taken in the lateral decubitus position under general anesthesia. Results Twenty-one eyes of 16 patients were included in the study. Median patient age at the time of surgery was 2.0 months [mode, 2.0; interquartile range (IQR), 1.6-2.6 months] and median follow-up time was 49.0 months (IQR, 33.0-99.5 months). The most common indication for surgery was Peters anomaly, in 16 eyes (76.2%). Ten eyes had additional intraocular surgeries posttransplant. The median ECDs prekeratoplasty and at last follow-up were 2958 cells per square millimeter (IQR, 2807-3205 cells/mm) and 1307 cells per square millimeter (IQR, 946-1613 cells/mm) respectively, reflecting a median endothelial cell loss of 59.2% (IQR, 44.3%-68.8%). Iris adhesions to the graft-host junction were strongly associated with low final ECD (P = 0.01). Conclusions Despite technical challenges and difficult postoperative care, pediatric keratoplasty is associated with a lower endothelial cell loss over time compared with that reported in adults after PKP.