Accuracy comparison of intraocular lens power calculation formulas in Post-LASIK myopic eyes

This study compared the accuracy of five intraocular lens (IOL) calculation methods (Barrett True-K, Haigis-L, Shammas-PL, EVO 2.0, and Kane) in post-LASIK myopic eyes. This study retrospectively analyzed 65 eyes that received phacoemulsification with IOL implantation after prior LASIK procedures for myopia correction. Five calculation methods (Barrett True-K, Haigis-L, Shammas-PL, EVO 2.0, and Kane) were applied to calculate anticipated refractive results. Prediction error (PE) and absolute prediction error (AE) were derived by comparing manifest refraction-derived spherical equivalents to formula-predicted values. Metrics included median PE (MedPE), median AE (MedAE), proportion of eyes achieving refractive outcomes within ±0.5 D and ±1.00 D, and statistical assessments evaluating axial length (AL)-related AE variations across formulas. EVO 2.0 formula demonstrated significantly lower prediction errors compared to PE than Shammas-PL ( P < 0.001) and Barrett True-K ( P = 0.001), while showing comparable performance to Haigis-L ( P = 0.052). Kane formula had higher PE than Barrett True-K ( P < 0.001), Haigis-L ( P < 0.001), Shammas-PL ( P < 0.001) and EVO 2.0 ( P < 0.001). Barrett True-K and Haigis-L formula achieved the lowest MedAE (median = 0.73), lower than Kane (median = 1.36, P < 0.001), but similar to EVO 2.0 (median = 0.76, P = 0.052) and Shammas-PL (median = 0.77, P = 0.118). Kane formula had a significantly fewer outcome within clinically acceptable ranges (±0.5D and ±1.0D) relative to other tested formulas ( P < 0.001). With increasing AL, AE of Barrett True-K and Kane formula showed a significant increase (Barrett True-K: r = 0.3497, P = 0.0043; Kane: r = 0.3084, P = 0.0124). EVO 2.0 demonstrated robust stability across diverse axial lengths (ALs), achieving superior predictive accuracy in post-LASIK IOL calculations compared to the Kane formula, while matching the performance of Barrett True-K, Haigis-L, and Shammas-PL methods. Considering accuracy, strategic integration of Barrett True-K and EVO 2.0 computational outcomes could optimize postoperative refractive results.