Astigmatism Prediction Error Of Two Oct Biometry Devices And Two Calculation Methods
Published 2024 - 42nd Congress of the ESCRS
Reference: PP10.02 | Type: Free paper | DOI: 10.82333/dn1y-4d98
Authors: Bjørn Gjerdrum* 1 , Kjell Gunnar Gundersen 1
1Ifocus Eye clinic,Haugesund,Norway
Purpose
To compare the accuracy of astigmatism predictions in IOL calculations using two different calculation methods and two different OCT biometers.
Setting
The study included results from two private eye clinics situated in Haugesund and Stavanger in Norway
Methods
This was a retrospective study of successful cataract surgery or refractive lens exchange. Preoperative biometry was performed with two different biometers: One SS-OCT biometer (Anterion, Heidelberg Engineering) and one combined reflection and SS-OCT biometer, (Eyestar 900, Haag-Streit). Calculations were performed with both devices using the built-in Barrett Toric calculator using both predicted (PPA) and measured posterior corneal astigmatism (MPA). The astigmatism prediction error (APE) was calculated as the magnitude of the vector differences between the achieved and the predicted refractive astigmatism. For patients with toric IOLs, the predicted astigmatism was recalculated using the actual postoperative toric alignments.
Results
The study included 95 eyes of 95 patients. The mean age was 61 years and 41% were female. Toric IOL´s were implanted in 40% of the eyes. Mean Follow-up time was 92 days (62 - 140). The mean uncorrected VA was -0,02 on the logMAR scale. The mean postoperative refractive spherical equivalent and astigmatism was 0.07 and -0,32, respectively. The mean APE varied from 0,35 to 0.36 for both calculations and both devices. The percentages of eyes with APE within 0.5 D was 81 and 82 for PPA and MPA, respectively with the Eyestar 900, and 77 and 79 for PPA and MPA, respectively with the Anterion. No statistically significant differences of means (p<0,4) or percentages of eyes within 0.5, 0.75 or 1 D of APE (p<0,1) were observed Conclusions
Conclusions
Both biometry decvices and both calculation methods showed acurate predictions with 77-85 % of eyes within a cylinder prediction error 0.5 D diopters of difference from the achieved postoperative astigmatism. No statistically significant differences in means or persentages of eyes within 0.5, 0.75 or 1 D or astigmatism.