Multiple Linear Regression Analysis of the Impact of Corneal Epithelial Thickness on Refractive Error Post Corneal Refractive Surgery

Purpose To evaluate the influence of corneal epithelial thickness (CET) on the deviation from the targeted refraction (refractive error [RE]) post corneal refractive surgery. Design Retrospective, single-center, cross-sectional study. Methods This study involved 211 eyes of 211 patients who previously underwent laser-assisted in situ keratomileusis (LASIK) or epipolis LASIK for the correction of myopia and myopic astigmatism from August 2000 to May 2014 at the Baptist Eye Institute, Kyoto, Japan, and who subsequently underwent examination of CET via optical coherence tomography imaging of 17 zones (central: n = 1 zone, paracentral: n = 8 zones, and midperipheral: n = 8 zones) within a 6.0-mm-diameter area of the central cornea from April 2014 to February 2015. The relationship between CET and RE was evaluated using Spearman coefficient in 5 parameters: CET in the (1) central, (2) paracentral, and (3) midperipheral zones, and the difference between the (4) central and paracentral zones and the (5) central and midperipheral zones. The influence of CET on RE was evaluated in the CET parameter that showed a significant correlation with RE using multiple liner regression analysis. Results The correlation coefficient with RE was (1) -0.238 (P < .01), (2) -0.172 (P < .05), (3) -0.002 (P = .98), (4) -0.186 (P < .01), and (5) -0.266 (P < .01), respectively. Multiple liner regression analysis revealed that the difference of mean CET between the central and midperipheral zones had a significant influence on RE (β coefficient = -0.028, P < .01). Conclusions The difference between CET at the central and midperipheral zones may play a role in the final RE post corneal refractive surgery. To evaluate the influence of corneal epithelial thickness (CET) on the deviation from the targeted refraction (refractive error [RE]) post corneal refractive surgery. Retrospective, single-center, cross-sectional study. This study involved 211 eyes of 211 patients who previously underwent laser-assisted in situ keratomileusis (LASIK) or epipolis LASIK for the correction of myopia and myopic astigmatism from August 2000 to May 2014 at the Baptist Eye Institute, Kyoto, Japan, and who subsequently underwent examination of CET via optical coherence tomography imaging of 17 zones (central: n = 1 zone, paracentral: n = 8 zones, and midperipheral: n = 8 zones) within a 6.0-mm-diameter area of the central cornea from April 2014 to February 2015. The relationship between CET and RE was evaluated using Spearman coefficient in 5 parameters: CET in the (1) central, (2) paracentral, and (3) midperipheral zones, and the difference between the (4) central and paracentral zones and the (5) central and midperipheral zones. The influence of CET on RE was evaluated in the CET parameter that showed a significant correlation with RE using multiple liner regression analysis. The correlation coefficient with RE was (1) -0.238 (P < .01), (2) -0.172 (P < .05), (3) -0.002 (P = .98), (4) -0.186 (P < .01), and (5) -0.266 (P < .01), respectively. Multiple liner regression analysis revealed that the difference of mean CET between the central and midperipheral zones had a significant influence on RE (β coefficient = -0.028, P < .01). The difference between CET at the central and midperipheral zones may play a role in the final RE post corneal refractive surgery.