Effect of surface photorefractive keratectomy and laser in situ keratomileusis on the corneal endothelium.

PURPOSE

To investigate endothelial cell loss in pairs of fresh human autopsy globes following high-diopter myopic photorefractive keratectomy (PRK) or laser in situ keratomileusis (LASIK).

SETTING

Center for Research on Ocular Therapeutics and Biodevices and Magill Laser Center for Vision Correction, Storm Eye Institute, Charleston, South Carolina, USA.

METHODS

In the first part of the study, 12 globes had either -10 diopters (D) multizone surface PRK or -10 D single-zone LASIK. In the second part, three groups of 5 globes each had -15 D, -20 D, or -25 D multizone-blend LASIK procedures. Fellow globes in both groups were used as untreated controls. Corneoscleral buttons were excised from all globes. Following 7 days in corneal organ culture, the endothelial surface was stained with two vital dyes: calcein-AM and ethidium homodimer. Fluorescence microscopy was used to obtain endothelial cell counts.

RESULTS

The mean dead cells per square millimeter (cells/mm2) were 0.94 in the -10 D PRK treated corneas compared with 0.91 in the fellow untreated eyes (P = 0.06(. The mean dead cells/mm2 in the -10 D single-zone LASIK-treated corneas and in the fellow untreated eyes were 0.61 (P = 0.88). The mean dead cells/mm2 in the -15 D, -20 D, and -25 D multizone-blend LASIK-treated corneas were 3.08, 2.33, and 5.55, respectively, compared with 3.49, 1.92, and 5.01 in the fellow untreated eyes (P = 0.276, P = 0.339, and P = 0.427, respectively). Dead cell counts for treated and control paired corneas were highly correlated in all treatment groups.

CONCLUSIONS

No significant endothelial cell loss occurred after -10 D PRK or LASIK corrections up to -25 D. Although this study has limitations that prevent direct extrapolation to the clinical situation, it does afford a comparable clinical correlate for endothelial cell toxicity following a typical excimer laser ablations.