A Regression Model for Correcting Intraocular Lens Power after Refractive Surgery Independent of Preoperative Data

Purpose To find a method of calculating intraocular lens (IOL) power that may be independent of preoperative data in eyes that have previously undergone myopic laser in situ keratomileusis (LASIK). Methods In 148 eyes of 75 patients, before and 6 months after LASIK, IOL power was calculated with SRK/T formula utilizing the spherical equivalent as the desired target refraction. Assuming that LASIK does not alter the crystalline lens refractive properties, IOL calculation error (CER) was estimated with this formula: CER = [pre-LASIK IOL power]/[post-LASIK IOL power]. Then the authors used postoperative biometry and Orbscan II corneal topography data in multiple regression models to find the best variables to predict the CER. Predicted amount of error which is calculated independent of preoperative data could be used to correct the post-LASIK calculated IOL: [corrected post-LASIK IOL power] = CER x [post-LASIK IOL power]. Results A regression model with these predictors was found: axial length in millimeters (L), radius of the anterior corneal surface best fitted sphere in millimeters divided by radius of the posterior corneal surface best fitted sphere in millimeters (AntBFS/PostBFS), corneal central 5 millimeters mean power in diopters divided by corneal central 3 millimeters mean power in diopters (mean 5 mm/mean 3 mm), the post-LASIK IOL power, and the post-LASIK simulated K reading. The model R square was 0.88. Conclusions There is correlation between post-LASIK biometry values and IOL power correction factor. This study presents a new model for further investigation.