Abstract

Objectives: To assess how differences in keratometric measurement principles affect intraocular lens (IOL) power calculation, we compared front-surface reflection-based (FSR) and anterior segment optical coherence tomography (AS-OCT)-derived keratometry (K) values. Methods: This retrospective study included 160 eyes of 160 Japanese patients who underwent cataract surgery with implantation of a CNA0T0 (Alcon) IOL between March 2023 and October 2025. FSR-based keratometry (OA-K, 2.5 mm averaged output ring) was obtained using the OA-2000 (Tomey), whereas AS-OCT-derived keratometry (CA-K, 2.5 mm single ring) and Fourier keratometry (CA-FK, 3.0 mm Fourier analysis zone) were obtained using CASIA2 (Tomey). Predicted refractions were calculated using Barrett Universal II (B-UII) and SRK/T formulas with each K value. Results: Mean K values were 44.14 ± 1.51 D (OA-K), 44.28 ± 1.51 D (CA-K), and 44.40 ± 1.55 D (CA-FK). With the B-UII, the mean absolute error (MAE) was lowest with OA-K (0.27 ± 0.24 D), followed by CA-K (0.28 ± 0.22 D) and CA-FK (0.31 ± 0.23 D). Corresponding MAEs with SRK/T were 0.30 ± 0.26 D, 0.31 ± 0.24 D, and 0.33 ± 0.28 D, respectively. Exploratory subgroup analyses suggested a tendency toward lower prediction accuracy in steep corneas. Differences among keratometric methods were smaller with SRK/T than with B-UII. Conclusions: AS-OCT-derived keratometry did not demonstrate superiority over conventional FSR-based keratometry for IOL power calculation. Further studies are warranted to determine whether AS-OCT-derived keratometry may provide clinical advantages under specific conditions.

IPC Classification

Keywords