Purpose
Improvements in LASIK outcomes will require advanced analyses to reveal correlations between preoperative measurements and systematic variations in outcomes. We postulate that pairwise and multivariate analyses could be used effectively to identify correctable systematic errors in LASIK outcomes.
Methods
This retrospective study analyzed the pre and post operative aberrometry and keratometry data on 120 eyes of 63 patients that underwent normal LASIK surgery using a 193 nm excimer laser equipped with iris registration. Pre- and post-surgery measurements were taken with a combined aberrometer and corneal topographer instrument. Wavefront measurements were expressed in Zernike polynomials and resized to 4 mm analysis pupil. JMP 8.0 software was used to analyze the third to fourth order Zernike coefficients for both pairwise and multivariate correlation and statistical significance.
Results
Statistically significant correlations were observed between pre operative and post operative wavefront coefficients. Figure 1 illustrates the pre to post operative multivariate correlations for the OD eyes; statistical significance for pairwise correlations are denoted with asterisks. Note that the correlation coefficients can exceed 40%. Both low and high order terms interact to affect the outcome. These data support the hypothesis that systematic errors in LASIK outcomes can be identified through these analyses.
Conclusions
That many of the correlations are statistically significant affords the hope that improvements in treatment planning can eliminate many of these errors thereby enhancing the quality of LASIK outcomes.
Keywords: 683 refractive surgery: LASIK •
684 refractive surgery: optical quality •
626 aberrations