Wound healing in rabbit corneas after photorefractive keratectomy and laser in situ keratomileusis

PURPOSE: To compare the wound-healing process in the rabbit cornea after photorefractive keratectomy (PRK) and laser in situ keratomileusis (LASIK) with the same refractive correction. SETTING: Department of Ophthalmology, Wakayama Medical University, Wakayama, Japan. METHODS: Adult albino rabbits (N = 24) were used. One eye of each animal had PRK or LASIK with the same refractive correction. Each animal was killed after an interval of up to 6 months. The expression pattern of corneal stromal injury-related molecules with the 2 treatments were compared. Paraffin sections of the cornea were processed immunohistochemically for alpha-smooth muscle actin (alpha-SMA), collagen type IV alpha1(IV)](2),alpha2(IV), and heat shock protein (HSP) 47 as well as other HSPs. Sections were also examined after hematoxylin and eosin or periodic acid-Schiff staining. RESULTS: Hematoxylin and eosin staining showed the central epithelium to be thick in PRK-treated corneas. The thick epithelium was restricted to the area around the corneal flap edge adhesion in LASIK-treated corneas at 3 months. Periodic acid-Schiff staining showed an absence of or interruption in the epithelial basement membrane in PRK-treated corneas for up to 6 months. Heat shock protein 47 was detected in keratocytes on day 3 but not after that in PRK-treated corneas. There was no difference in the expression of other HSPs. Alpha-smooth muscle actin was expressed in keratocytes repopulated in the central anterior cornea of PRK-treated corneas at 28 days. Keratocytes with immunoreactivity for these 2 proteins were not seen in LASIK-treated corneas. Collagen IV alpha1(IV)](2),alpha2(IV) was not detected in either group of corneas. The central epithelium became transiently thicker in PRK-treated corneas. CONCLUSION: Keratocyte responses to laser stromal ablation were more marked in corneas treated with PRK than in those treated with LASIK.