Accuracy of Intraocular Lens Power Prediction Using 9 Formula

Purpose: To compare the accuracy of nine intraocular lens (IOL) power calculation formulas.Methods: This study is retrospective consecutive case series. A total of 228 eyes of 228 patients who underwent uncomplicated cataract surgery between October 2015 and March 2021 were included. The accuracy of nine IOL calculation formulas (Kane, Emmetropia verifying optical version, Hill-radial basis function, Olsen, Barrett Universal II [Barrett II], Haigis, Holladay, Hoffer, and SRK/T) was compared and analyzed using differences between the predicted refractive power and actual refractive power at 2 months after cataract surgery. Mean prediction error (ME), mean absolute error (MAE), median absolute error (MedAE), and the percentage of eyes that were within ± 0.50 diopters (D), ± 0.75 D, and ± 1.00 D of the target refraction were calculated for each formula.Results: For all eyes, Haigis formula was the most accurate in MAE, followed by Barrett II formula. Comparing the probability of MAE within 0.50 D, Haigis formula showed the highest accuracy. A total of 20 eyes of 20 patients had more than 25 mm axial length. Regarding MAE, Hoffer Q was the most accurate, followed by the Olsen formula. Comparing the probability of MAE within 0.50 D, the Hoffer Q formula was the most accurate. A total of 24 eyes of 24 patients had more than 46 D keratometry. Regarding MAE and MedAE, Haigis was the most accurate, followed by the Barrett II formula. Comparing the probability of MAE within 0.5 D, Holladay formula was the most accurate.Conclusions: Barrett II and Haigis formulas showed greater accuracy for all patients when we determined IOL power before cataract surgery. Hoffer Q, Olsen, and Barrett II formulas showed higher accuracy for long axial length. And Haigis and Barrett II were the most accurate for steep keratometry.