Evaluating predictive accuracy of intraocular lens power calculation formulas in high axial myopia: A comparative analysis

PURPOSE: To compare the precision of six intraocular lens (IOL) power calculation formulas-Barrett Universal II, SRK/T, Haigis, Holladay I, Holladay II, and Hoffer Q-among myopic patients undergoing cataract surgery. METHODS: This retrospective study included 164 eyes from 131 patients with an axial length (AL) greater than 26.0 mm who underwent uncomplicated cataract surgery with monofocal IOL implantation at a tertiary referral center between 2018 and 2021 Several biometric data points were recorded using the IOLMaster 700. Additionally, refractive prediction error was observed and defined as the difference between predicted and postoperative spherical equivalents. Formula performance was assessed using mean absolute error (MAE), mean numerical error, median absolute error, and proportion of eyes achieving predefined refractive accuracy (±0.25 D, ±0.50 D, ±1.00 D, and ± 2.00 D). RESULTS: Barrett Universal II showed the lowest calculation of MAE (0.54 ± 0.71 D), followed by SRK/T (0.57 ± 0.75 D) and Haigis (0.61 ± 0.69 D). Barrett Universal II demonstrated significant predictive accuracy compared with Holladay I, Holladay II, and Hoffer Q ( P < 0.05), while no statistically significant differences were observed when compared to SRK/T and Haigis. Besides, it resulted in the highest proportions of eyes within ± 0.25 D (36.6%), ±0.50 D (59.1%), ±1.00 D (88.4%), and ± 2.00 D (98.8%). The absolute prediction error increased with AL ( P < 0.05), whereas there was no significant correlation with keratometry. CONCLUSION: In highly myopic eyes, Barrett Universal II showed numerically lower refractive prediction errors, with overall accuracy comparable to SRK/T and Haigis. No formula demonstrated consistent statistical superiority, indicating broadly equivalent performance among the leading formulas in this population.