Femtosecond laser and microkeratome corneal flaps: comparison of stromal wound healing and inflammation

Abstract

Purpose: To examine early postoperative wound healing in rabbit corneas that had LASIK flaps formed with three different models (15 KHz, 30 KhZ, and 60 KHz) of a femtosecond laser compared with flaps formed with a microkeratome.

Methods: Thirty-nine rabbit eyes were randomized to receive either no surgery or corneal flaps formed with one of the lasers or the microkeratome. Sixteen eyes also had lamellar cuts with no side cuts with the 30 KHz laser. Animals were sacrificed and corneas processed as frozen sections or fixed for transmission electron microscopy. Frozen sections were evaluated with the TUNEL assay to detect apoptosis, immunocytochemistry for Ki67 to detect cell mitosis, and immunocytochemistry for CD11b to detect mononuclear cells.

Results: Rabbit corneas that had flaps formed with the 15 KHz laser had significantly more stromal cell death, greater stromal cell proliferation, and greater monocyte influx in the central and peripheral comea at 24 hours after surgery than corneas that had flaps formed with the 30 KHz or 60 KHz laser or the microkeratome. Results of the 60 KHz laser and microkeratome were not significantly different for any of the parameters at 24 hours, except for mitotic stromal cells at the flap margin. Transmission electron microscopy revealed that the primary mode of stromal cell death at 24 hours after laser ablation was necrosis.

Conclusions: Stromal cell necrosis associated with femtosecond laser flap formation likely contributes to greater inflammation after LASIK performed with the femtosecond laser, especially with higher energy levels that result in greater keratocyte cell death.

Figure 1

Terminal deoxyribonucleotidyl transferase–mediated dUTP-digoxigenin nick end labeling (TUNEL) assay 24 hours after surgery of the central cornea from rabbits that had flap formation with the A) Hansatome microkeratome (Bausch & Lomb, Rochester, NY), B) the 15 KHz Model II femtosecond laser (IntraLase, Irvine, Calif), and C) the 30 KHz Model II femtosecond laser (IntraLase). Cell nuclei are stained blue with 4’,6 diamidino-2-phenylindole and TUNEL-positive cells are stained red (arrows indicate some TUNEL-positive cells in each panel). The level of TUNEL-positive cells triggered by the 60 KHz Model II femtosecond laser (not shown) was similar to that noted with the microkeratome. Original magnification ×400.

Figure 2

Terminal deoxyribonucleotidyl transferase–mediated dUTP-digoxigenin nick end labeling (TUNEL) assay and transmission electron microscopy in corneas where an intrastromal cut was produced with the 30 KHz Model II femtosecond laser (IntraLase, Irvine, Calif) without side cuts and a higher than normal energy for this model of 2.7 μJ. (Fig 2A: Large arrowhead indicates end of the stromal laser cut; arrows show the lamellar cut; and small arrowheads show keratocytes.) Thus, there was no epithelial injury produced by the laser ablation. Although positive cells (Fig 2B arrows show TUNEL and stromal cells; Fig 2C, D, and E arrows and arrowheads show necrotic cellular debris) were detected at the site of femtosecond laser ablation using the TUNEL histological assay (Fig 2A and 2B original magnification ×400), only necrotic cells with random disruption of cellular organelles without condensed chromatin or cell shrinkage were detected with transmission electron microscopy (Fig 2 C, D and E, original magnification ×5000).

Figure 3

Mitotic Ki67-stained cells at 24 hours after surgery of the peripheral cornea at and central to the flap edge in rabbits that had flaps formed with A) the Hansatome microkeratome (Bausch & Lomb, Rochester, NY), B) the 15 KHz Model II femtosecond laser (IntraLase, Irvine, Calif), and C) the 30 KHz Model II femtosecond laser (IntraLase). Cell nuclei are stained blue with 4’,6 diamidino-2-phenylindole and Ki67-positive cells are stained red (arrows indicatedsome in each panel). The arrowheads in each panel are the epithelial plugs at the flap margin. In the particular section for the microkeratome shown here, there was no stromal cell proliferation immediately joining the epithelial plug. Other microkeratome specimens had some mitotic cells near the epithelial plug, but never to the extent noted with either the 15 KHz or the 30 KHz laser. Counts were performed at the microscope and care was taken to include only stromal cells and not epithelial cells within the plugs. Images of mitotic Ki67-positive cells in the peripheral cornea with the 60 KHz Model II femtosecond laser are not shown, but in general tended to be similar to those with the 30 KHz laser, although some corneas showed far less Ki67-stained cells, similar to the level noted with the microkeratome (not shown). Original magnification ×400.

Figure 4

Monocytes expressing CD11b detected at 24 hours after surgery with the A) Hansatome microkeratome (Bausch & Lomb, Rochester, NY), B) the 15 KHz Model II femtosecond laser (IntraLase, Irvine, Calif), and C) the 30 KHz Model II femtosecond laser (IntraLase). There was typically little monocyte infiltration in the center of corneas (arrowheads) with flaps produced with the microkeratome. Monocytes were routinely detected at the flap margins (arrows) near the side cut through the epithelium in the microkeratome group. Greater numbers of monocytes were detected in both the central (arrowheads) and peripheral (arrows) cornea after flap formation with either of the femtosecond lasers, although the mean was significantly greater with the 15 KHz laser. The cornea treated with the 30 KHz laser shown here C) had the highest level of monocyte infiltration of any of the corneas in that group, whereas the 15 KHz cornea shown here B) was average for that group. The levels of monocyte infiltration in corneas that had flaps formed with the 60 KHz laser were similar to those in corneas that had flaps formed with the 30 KHz laser, although some had lower levels similar to those in the microkeratome group (not shown). Original magnification ×200.