Corneal backscatter after penetrating keratoplasty

Abstract

Abstract Purpose: To determine changes in corneal backscatter in clear grafts after penetrating keratoplasty (PK) and in grafts that had failed from late endothelial failure (LEF). Methods: Sixty‐one grafts, deemed clear by slit‐lamp examination, in 46 patients and 12 grafts with LEF in 11 patients were examined at 1 to 30 years after PK. Backscatter was measured from the anterior, middle and posterior thirds of the cornea from a digitized image of a high magnification slit‐beam through the center of the cornea. Clear grafts were compared to grafts with LEF and to 41 normal (unoperated) corneas of 41 subjects by using unpaired t‐tests. Photopic low contrast visual acuity (LCVA) was measured in 32 of the clear grafts. Generalized estimating equation models were used to account for correlation between fellow eyes of the same patient. Results: Backscatter was increased by 80%, 49% and 34% in grafts with LEF compared to clear grafts in the anterior (P< 0.001), middle (P< 0.001) and posterior (P= 0.01) thirds of the cornea, respectively. Backscatter was increased by 16%, 18% and 29% in clear grafts compared to normal corneas in the anterior (P< 0.001), middle (P< 0.001) and posterior (P< 0.001) thirds of the cornea, respectively. Backscatter from clear grafts correlated with time after keratoplasty in the anterior (r= 0.23, P= 0.005), middle (r=0.21, P= 0.006) and posterior (r=0.32, P= 0.003) thirds of the cornea. LCVA correlated with backscatter in the anterior (r= ‐0.63, P< 0.001), middle (r= ‐0.56, P< 0.001) and posterior (r= ‐0.51, P= 0.003) thirds of the cornea. Conclusions: Backscatter from penetrating corneal grafts is higher than normal, and substantially higher in grafts with LEF. Backscatter increases with time after keratoplasty, and increased backscatter is associated with decreased LCVA.