Evaluation of corneal stromal changes in vivo after laser in situ keratomileusis with confocal microscopy

Purpose: To assess by in vivo confocal microscopy the modifications of the corneal stroma after laser in situ keratomileusis (LASIK) for myopia. Design: Nonrandomized comparative (self-control led) trial. Participants: Sixteen eyes of 13 patients were examined before surgery and at days 8, 30, and 90, and 9 eyes were examined at 6 months postoperatively using an in vivo confocal microscope. Testing/Intervention: Stromal morphologic changes, keratocyte density, flap thickness, and subclinical haze were evaluated and compared at different t ime points. LASIK was performed with a Flapmaker microkeratome (Solan Ophth almic products, Jacksonville, FL) and a Lasersight LSX excimer laser (Laser Sight Technologies Inc., Winter Park, FL). Main Outcome Measure: Confocal microscopy results. Results: Microfolds at the Bowman's layer were found in most eyes, as well as variable reflectivity particles (pa) located at the interface level in a ll eyes examined postoperatively. The density of these particles significan tly decreased with time with, respectively, 504 +/- 101 pa/mm(2) at day 8 a nd 380 +/- 111 pa/mm(2) at day 30 (P = 0.003), 332 +/- 100 pa/mm(2) at mont h 3 and 312 +/- 40 pa/mm(2) at month 6. The mean flap and the activated-cel ls area thicknesses were, respectively, 102 +/- 26 mum and 61 +/- 19 mum an d showed significant negative correlation (P < 0.0001). The intensity of th e added peak (47.3 <mu>m 8.6%), corresponding to the subclinical haze, real ized by Z-scan measure, was also negatively correlated with flap thickness (P = 0.01). Keratocyte (k) density quantified in the posterior stroma signi ficantly increased from day 0 (480 +/- 67 k/mm(2)) to day 8 (701 +/- 41 k/m m(2), p < 0.0001 compared with day 0) and day 30 (917 +/- 143 k/mm(2), p = 0.0006, compared with day 0) but significantly decreased at 3 months postop eratively (597 +/- 56 k/mm(2), p < 0.0001 compared with day 30) to reach th e initial level at month 6 (502 +/- 41 k/mm(2), nonsignificant compared wit h day 0). There was no correlation between preoperative or postoperative sp herical equivalent and the density of particles, keratocytes, and the haze intensity. Conclusions: This study confirms the presence of microfolds and particles a t the interface level, as well as subclinical impairment. Evaluation of ker atocyte density constitutes a major contribution of confocal microscopy tow ard an understanding of the keratocyte response to corneal wound healing af ter corneal refractive surgery. Moreover, flap thickness seems to be involv ed in the postoperative cellular activation with a higher response when thi n. Ophthalmology 2001;108:1744-1750 (C) 2001 by the American Academy of Oph thalmology.