Immunohistochemical Findings After LASIK Confirm In Vitro LASIK Model.

In recent years, laser in situ keratomileusis (LASIK) became an established surgical procedure to correct refractive error. Numerous patients have undergone this treatment that can be considered a safe procedure within biologic limits. However, one aspect still in discussion is the wound-healing process in the created interface that leads to an easily removable flap even years after treatment.

Although we have learned quite a lot from experimental animal models and in vivo confocal studies, histologic studies on human LASIK-treated eyes that are required to study and better understand the complete tissue reaction in vivo so far are limited to case reports of corneas, most of them with postoperative complications. They demonstrated absence of bridging collagen fibrils and cells between the flap undersurface and the stromal bed and epithelial ingrowth. To date there are only a few case reports of corneas with prior uncomplicated LASIK: in one case, epithelial plugs under the flap margin, regeneration of nerve fibers in the anterior stroma, and a restricted expression of fibronectin and tenascin at the wound area were described. By comparing LASIK wound healing after 3 and 20 months, Anderson et al could demonstrate a reduction of reactive wound repair like epithelial ingrowth, reactive keratocytes at the wound margin, and irregular collagen fibrils in the wound bed. Recently, Mootha and colleagues showed vacuolization and pyknosis of keratocytes as a consistent histopathologic finding in a series of donor corneas with prior LASIK.

To gain further insight in morphologic changes of human eyes after LASIK surgery, we recently established an in vitro LASIK model using human donor eyes. This in vitro model showed similarities to studies of animals treated with LASIK, but has not yet been confirmed in the human in vivo situation. To test the validity of the human in vitro model, one donor with a complete history and follow-up of a LASIK procedure was obtained from the eye bank of the Department of Ophthalmology in Munich. The LASIK treatment in this donor showed no complications during the observation period of 8 months in vivo.

DISCUSSION

Immunohistochemical studies of donor corneas with prior successful LASIK surgery revealed a similar staining pattern as cultured corneas in the in vitro LASIK model presented recently. Staining for fibronectin in the donor corneas was consistent along the complete incision line, whereas staining for collagen type I and laminin did not show alterations in the central interface. The difference in fibronectin staining of the LASIK interface in human LASIK eyes with clinical complications, where the presence of fibronectin was restricted to regions of epithelial cell ingrowth, and in rabbit LASIK eyes might be due to differences in the sensitivity of the antibody or in the method of sectioning (cryosections versus paraffin sections).

Consistent with other studies of human LASIK eyes, no staining for collagen type III could be detected in the central region of the LASIK interface. However, similar to the rabbit LASIK eyes, collagen type III was faintly present at the peripheral incision side. In contrast to the donor corneas after LASIK, in our previously published human in vitro LASIK model, collagen type III staining was present in the central area of the human flap. This might most likely be explained due to a demasking of collagen type III following the culturing process immediately after LASIK procedure.

The lack of pronounced morphologic changes in the central area of the LASIK interface, which only showed little accumulation of fibronectin, supports the hypothesis of reduced wound-healing reactions after performing this surgical procedure. Only at the rim zone of the incision, scar tissue formation can appear and might form an incomplete fixation zone for the corneal flap. Due to this impaired healing process, even years after the LASIK procedure, a corneal flap displacement can occur.

In summary, our histologic findings confirm the well-known clinical phenomenon that wound-healing reactions are marginal after uncomplicated LASIK treatment. Because animal models often have limitations in comparison with the human situation and some differences exist between human and rabbit LASIK treated eyes, similar findings in the in vivo situation after LASIK treatment and in the recently introduced human in vitro model strengthen the reliability of the latter model for further studies concerning LASIK.

Conclusion:

Histologic findings in donor corneas with prior LASIK treatment confirm histologic observations in a recently introduced human organ culture LASIK model. This strengthens the reliability of the latter LASIK model for further studies concerning wound healing after LASIK surgery.