Keratocyte Density After Penetrating Keratoplasty and in Late Endothelial Failure

Abstract

Purpose:: To determine changes in keratocyte density after penetrating keratoplasty (PK), and to compare keratocyte density in grafts with late endothelial failure to keratocyte density in clear grafts.

Methods:: 58 clear corneal grafts of 45 patients were examined by using a Tandem Scanning confocal microscope at 1 to 30 years after PK. Central keratocyte density was measured in 5 anterior-posterior stromal layers by using a custom program that objectively identified bright objects (presumed to be keratocyte nuclei) in confocal images. Measurements were repeated in 16 clear grafts (13 patients) after several years. Cell density was also measured in 44 normal corneas of 44 patients and in 19 grafts of 17 patients with late endothelial failure. Keratocyte densities were compared between clear grafts, normal corneas, and grafts with late endothelial failure. We also compared the number of keratocytes in a full-thickness column of stroma with frontal area of 1 mm².

Results:: Mean keratocyte density for the full thickness stroma of 58 clear grafts (22,134 ± 3,447 cells/mm³, mean ± SD) was lower than in 44 normal corneas (26,696 ± 3,046 cells/mm³, P< 0.001, unpaired t-test). The number of keratocytes in a column of full-thickness stroma in clear grafts (10,495 ± 1,477 cells) was also lower than in normal corneas (11,276 ± 1,217 cells, P= 0.005, unpaired t-test). Time after PK correlated with keratocyte density in the middle third (r= -0.46, P< 0.001), posterior 67-90% (r= -0.42, P= 0.001), and posterior 10% (r= -0.31, P= 0.02) of the stroma (n=58). In 16 clear grafts examined twice 57 ± 8 months (range, 43-68 months) apart, keratocyte density decreased in the middle third (P< 0.001), posterior 67-90% (P< 0.001) and posterior 10% (P= 0.003) of the stroma (paired t-tests), and the annual rate of keratocyte loss was 1.6 ± 2.2%. In 9 grafts with late endothelial failure, neither mean keratocyte density (21,276 ± 3,223 cells/mm³) nor the number of keratocytes in a column of full-thickness stroma (10,390 ± 1,582 cells) differed from those in clear grafts (P> 0.49, unpaired t-tests; minimum detectable differences were 3,500 cells/mm³ and 1,520 cells respectively, α= 0.05, ß= 0.20). Cell density could not be measured in 10 grafts with late endothelial failure because images were hazy and cells could not be identified.

Conclusions:: Central keratocyte density and the number of keratocytes are decreased in penetrating grafts compared to normal corneas. The annual rate of keratocyte loss after PK is greater than in normal corneas. Grafts with late endothelial failure do not have increased loss of keratocytes compared to clear grafts.

Keywords: cornea: stroma and keratocytes • transplantation • microscopy: confocal/tunneling