Accuracy and refractive adjustment of intraocular lens calculation formulas for primary intraocular lens fixation using the Yamane technique

PURPOSE: To evaluate the prediction accuracy of intraocular lens (IOL) calculation formulas in primary IOL fixation using the Yamane technique and explore adjustment strategies to improve the refractive outcomes. SETTING: Zhongshan Ophthalmic Center, Guangzhou, China. DESIGN: Retrospective case series. METHODS: 50 eyes from 50 patients with inadequate capsular support who underwent lens extraction and primary Yamane IOL fixation were reviewed. The predictive accuracy of 7 formulas, Barrett Universal II (BU II), Emmetropia Verifying Optical (EVO), Kane, SRK/T, Holladay 1, Hoffer Q, and Haigis, was evaluated with or without lens constant optimization. RESULTS: Before lens constants optimization, SRK/T, Holladay 1, Hoffer Q, BU II, and EVO formulas exhibited statistically significant hyperopic mean prediction errors (PEs) (range 0.26 to 0.44 diopters [D], all P < .05), whereas the Haigis and Kane formulas showed relatively better accuracy without significant hyperopic shift. The mean absolute error (MAE) of all formulas ranged from 0.47 to 0.65 D, and the percentage of eyes with PE with ± 0.50 D ranged from 40% to 60%. After lens constants optimization, the BU II and SRK/T formulas exhibited the lowest MAE (0.43 and 0.47, respectively), with the highest percentage of PE within ± 0.50 D (64% and 64%, respectively). CONCLUSIONS: Before lens constant optimization, BU II, EVO, SRK/T, Holladay 1, and Hoffer Q all produced significantly hyperopic PEs in primary Yamane IOL fixation. After lens constants optimization, the BU II and SRK/T displayed better performance. To enhance refractive outcomes in these patients, a slightly myopia target refraction or optimization of the lens constants was recommended.