ESCRS - PP13.17 - Comparison Of 11 Intraocular Lens Power Calculation Methods In Short Eyes

Comparison Of 11 Intraocular Lens Power Calculation Methods In Short Eyes

Published 2022 - 40th Congress of the ESCRS

Reference: PP13.17 | Type: Free paper | DOI: 10.82333/5pwn-b860

Authors: Bernardo Feijóo* 1 , Filomena Ribeiro 2 , Francisco Castro Alonso 3 , Tiago Ferreira 4 , Sylvia Gaspar 4

1Centro da Visão,Hospital da Luz Lisboa,Lisboa,Portugal, 2Centro da Visão,Hospital da Luz Lisboa,Lisboa,Portugal;Visual Sciences Research Center,University of Lisbon,Lisbon,Portugal;Medical School,University of Lisbon,Lisbon,Portugal, 3Hospital de Alcañiz,Alcañiz,Spain, 4Centro da Visão,Hospital da Luz Lisboa,Lisbon,Portugal

Purpose

To evaluate the accuracy of 11 intraocular lens (IOL) calculation methodologies in short eyes: SRK-T, Holladay 1, Haigis, Hoffer-Q, Barret Universal II, Olsen, RBF, Panacea, EVO, Kane and PEARL.

Setting

Centro da Visão, Hospital da Luz Lisboa, Portugal and Department of Ophthalmology, Hospital de Alcaniz, Spain

Methods

Retrospective case series comparison. We enrolled 178 eyes with axial length equal or inferior to 22 mm. All of them had had routine ocular examination and optic biometry (Lenstar) preoperatively and then underwent phacoemulsification with implantation of a monofocal IOL. We compared the postoperative residual subjective spherical equivalent with the predicted refractive outcome using each methodology.

Results

The mean error was adjusted to zero. The different performances of each methodology were ranked according to standard deviation (SD) of the mean refractive error. The EVO formula had the lowest SD, the lowest absolute median error (0,24D) and the largest proportion of eyes with an absolute error equal or lower than 0,50 D (77,9%). Barret formula had similar results with the same proportion of eyes reaching 0,50 D of the predicted error. There was a statistically significant difference between the Barret formula (absolute median error 0,24 D) and SRK/T (absolute median error 0,31D).

Conclusions

Our results demonstrate that refractive prediction of current IOL power calculation methodologies in short eyes are still inferior to the described for samples with larger axial lengths. Barret formula offers a better performance than SRK/T.