Accuracy of new intraocular lens calculation formulas in Chinese eyes with short axial lengths.

Abstract

To compare the measurement accuracy of new/updated intraocular lens (IOL) power calculation methods, namely, Kane, Emmetropia Verifying Optical (EVO), with existing methods (Barrett Universal II, Olsen, Haigis, Hoffer Q, Holladay 1, SRK/T) in Chinese eyes with axial lengths ≤ 22.5 mm. The study included data from patients who underwent uneventful cataract surgery with the insertion of ZCB00 IOL. Refractive prediction errors were determined by calculating the difference between postoperative refraction and the predicted refraction using each formula. Various parameters were evaluated, including mean prediction error (ME), mean absolute error (MAE), median absolute error (MedAE), and the percentage of eyes with prediction errors (PE) within different ranges. The study enrolled 38 eyes of 38 patients, and the Barrett Universal II formula demonstrated the lowest MAE and MedAE among the tested formulas. Post hoc analysis using Wilcoxon signed-rank pairwise comparisons for non-parametric samples with Bonferroni correction revealed no significant difference in postoperative refractive prediction among all the formulas (P > 0.05). The percentage of eyes with PE within ± 0.5 D was as follows: Barrett Universal II, 81.58%; Haigis, 78.95%; EVO, 76.32%; Olsen, 76.32%; Holladay I, 73.68%; SRK/T, 71.05%; Kane, 68.42%; and Hoffer Q, 65.79%. The Barrett Universal II formula was more accurate than the other formulas for Chinese eyes with AL ≤ 22.5 mm.