Comparison of wound healing after photorefractive keratectomy and laser in situ keratomileusis in rabbits

Purpose: To evaluate and compare the corneal wound-healing process after photorefractive keratectomy (PRK) and laser in situ keratomileusis (LASIK). Setting: Kangnam St. Mary's Hospital, Seoul, Korea. Methods: Two surgical procedures, PRK with the VISX Star excimer laser and LASIK with a MicroTech microkeratome, were performed in 24 rabbit eyes. In the PRK group (n = 12 eyes), the rabbit cornea was treated with a 20 mu m ablation. In the LASIK group (n = 12 eyes), a 20 mu m laser ablation was performed after a 150 mu m thick hinged corneal flap had been made. During both procedures, dichlorotriazinyl aminofluorescien (DTAF) dye was applied to the ablated stromal bed; in the LASIK group, the stromal side of the corneal flap was also stained with DTAF to differentiate regenerated collagen from normal stromal tissue. Corneal wound healing was evaluated postoperatively at 1, 4, 8, and 12 weeks using light, electron, and fluorescence microscopy. The amount of regenerated stromal tissue and the number of keratocytes were analyzed by an image-analysis system. Results: In the PRK group, epithelial migration and regeneration were observed in the ablated area without any stromal regeneration 1 week postoperatively, However, newly regenerated, irregularly arranged stromal collagen, with epithelial hyperplasia in the ablated area, was observed 4 to 12 weeks postoperatively by light and fluorescence microscopy. The number of keratocytes in the surgical area was also increased. In ultrastructural observation using an electron microscope, the shape of keratocytes in the ablated area was changed, and the number of rough and smooth endoplasmic reticuli, ribosomes, mitochondria, and electron-dense vesicles in the cytoplasm were increased, suggesting that the cells were activated. In the LASIK group, there was no observed regenerated collagen between the corneal flap and the ablated stromal bed except in the wound margin. Lamellated, parallel collagen fibers in the cornealstroma were not disturbed. However, in the wound margin, corneal epithelial ingrowth between the flap and the stromal bed was observed, as was some regenerated stromal tissue. The amount of regenerated stromal tissue and the number of keratocytes in the wound area were statistically smaller than those in the PRK group (P <.05). Observation by electron microscopy showed no activated keratocytes, unlike in the PRK group. The collagen fibers in the wound area were parallel. Conclusion: Stromal wound healing in the LASIK group was minimal compared with that in the PRK group, except in the wound margin. These results may support the clinical findings of less corneal haze in the human cornea after LASIK. J Cataract Refract Surg 1999; 25:842-850 (C) 1999 ASCRS and ESCRS.