Résumé
Purpose: to examine biometric factors inducing a prediction error related to lens position in patients undergoing cataract surgery with the TECNIS Synergy intraocular lens (IOL). Setting: department of Ophthalmology, Montpellier University Hospital, Montpellier, France. Design: consecutive retrospective case series. Methods: we included all patients with implantation of the TECNIS Synergy IOL during phacoemulsification from January 2021 to July 2024. We collected clinical data (age, sex) and biometry data such as axial length, anterior chamber depth, lens thickness, and white-to-white distance. Patients underwent postoperative refraction assessment and biometry (IOLMaster 700, ZEISS) pre-operatively and 1 month post-operatively. We assessed the mean prediction error (MPE) with four formulas (Barret Universal II [BUII], SRK/T, Hoffer Q [HQ], post-operative spherical equivalent prediction using artificial intelligence and linear algorithms by Debellemaniére, Gatinel, and Saad [PEARL DGS]) and number of eyes within ±0.25 diopters (D), ±0.5 D, ±0.75 D, ±1.0 D, and ±2.0 D of target refraction. Results: we included 17 eyes from 17 patients; the mean age of patients was 62.2±6.9 years. The MPE was assessed as myopic with all formulas: -0.60±0.61 D (BUII); -0.65±0.64 D (SRK/T), -0.68±0.62 D (HQ), and -0.68±0.48 D (PEARL DGS) (P=0.077). Many eyes (58.8%) were within ±0.75 D with BUII, followed by SRK/T (52.9%), PEARL DGS (47.0%), and HQ (47.0%). The mean actual lens position (ALP) was 5.5±0.3 mm and mean C Constant was 0.53±0.07. With all formulas, the MPE was not correlated with mean ALP. The C constant was correlated with the MPE with the PEARL DGS formula only (r=-0.5; R2=0.252; p value=0.04). Conclusion: all formulas exhibited comparable performance, with a myopic shift that would be interesting to consider in our practice. Furthermore, with the BUII, the greatest number of eyes was within ±0.75 D. However, the difference between the formulas was not statistically significant. Similarly, mean ALP was not correlated with its refraction errors, regardless of formula used. In contrast, the C constant was negatively correlated with the MPE with only the PEARL DGS formula.