ESCRS - PP08.04 - Phaco-Descemet Stripping Endothelial Keratoplasty: Evaluating The Accuracy Of Intraocular Lens Power Formulas

Phaco-Descemet Stripping Endothelial Keratoplasty: Evaluating The Accuracy Of Intraocular Lens Power Formulas

Published 2026 - 44th Congress of the ESCRS

Reference: PP08.04 | Type: Free paper | DOI: 10.82333/987z-qx53

Authors: Dr Ricardo Costa-Gertrudes* 1 , Dr Nuno Rodrigues Alves 2 , Dr Catarina Barão 2 , Dr Sara Crisóstomo 2 , Dr João Feijão 2 , Dr Vitor Maduro 2 , Dr Nuno Filipe Alves 2 , Dr Diogo Hipólito-Fernandes 2

1Ophthalmology,ULS São José,Lisbon,Portugal;Faculty of Medicine,University of Lisbon,Lisbon,Portugal, 2Ophthalmology,ULS São José,Lisbon,Portugal

Purpose

To assess and compare the accuracy of six modern intraocular lens (IOL) power calculation formulas in patients submitted to combined phaco-Descemet Stripping Endothelial Keratoplasty (DSAEK)(triple-DSAEK).

Setting

Department of Ophthalmology, Unidade Local de Saúde de São José, Lisbon, Portugal

Methods

Retrospective study including patients submitted to uneventful triple-DSAEK with the implantation of hydrophobic monofocal IOL in-the-bag (Alcon Acrysof®️ AU00T0). All patients underwent optical biometry (Haag-Streit Lenstar LS-900). Refraction prediction error (PE) of each formula was calculated as the difference between the subjective refraction spherical equivalent (SE) and the targeted SE of each formula from the ESCRS IOL Calculator (Barrett Universal II, EVO 2.0, Hill-RBF, Hoffer QST, Kane and PEARL‐DGS). Outcome measures included the mean prediction error (ME), mean (MAE) and median (MedAE) absolute errors, and the percentage of eyes within ± 0.25D, ± 0.50D and ± 1.00D.

Results

Twenty-five eyes with a mean axial-length of 23.01 ± 1.41 mm (mean [SD] age, 76.21 [9.9] years; 62.5% females) were included. Kane Formula had the lowest MAE (0.618) and MedAE (0.450), followed by EVO 2.0 (MAE 0.628) and Hoffer QST (MAE 0.631). The difference in formulas absolute error was not significative (p = 0.931). In all formulas, a slight hypermetropic shift of 0.2-0.3D was observed, but it was not significantly different from zero (p > 0.05). The Kane formula yielded a prediction error within ± 0.25D, ± 0.50D, and ± 1.00D in 20.8%, 58.3%, and 79.2% of cases, respectively.

Conclusions

In patients undergoing combined phaco-DSAEK, our investigation revealed a non-significant hypermetropic shift of 0.2-0.3D across all studied formulas. The Kane Formula exhibited superior predictive accuracy, evident through its lowest mean absolute error (MAE), lowest median absolute error (MedAE), and the highest percentage of eyes with a PE within ±  0.50D. These results substantiate the effectiveness of advanced IOL calculation formulas, particularly the Kane Formula, in enhancing precision and optimizing outcomes for combined phaco-DSAEK procedures.