To Measure And Predict Post-Operative Capsular Bag Stability Of A Novel Glistening-Free Hydrophobic Single-Piece Intraocular Lens Using A Machine Learning Algorithm

Published 2024 - 42nd Congress of the ESCRS

Reference: PP13.09 | Type: Free paper | DOI: 10.82333/pnnb-jf50

Authors: Klemens Waser* ¹ , Klaus Strassmair ¹ , Peter Laubichler ¹ , Siegfried Mariacher ¹ , Haidar Khalil ¹ , Leon Pomberger ¹ , Matthias Bolz ¹ , Nino Hirnschall ¹

¹Ophthalmology,Kepler University Clinic Linz,Linz,Austria;Johannes Kepler University Linz,Linz,Austria

Purpose

In the course of cataract surgery, an artificial lens is implanted instead of the crystalline lens. The post-operative outcome significantly depends on the effective post-operative lens position and thus the post-operative anterior chamber depth (ACD). However, this changes over the course of the post-operative phase. The aim of this study is, on one hand, to measure the post-operative IOL shift and its impact on post-operative refraction, and on the other hand, to predict the IOL shift from pre-operative biometric data using a bootstrapping model and a machine learning algorithm.

Setting

Patients scheduled for cataract surgery at Kepler University Clinic Linz, Austria were included. A positive ethical approval was obtained (EK Nr. 1044/2023) from the Ethics Committee of the Medical University Linz, JKU, and the study adheres to the principles outlined in the Declaration of Helsinki.

Methods

The underlying study is a prospective, single-center, non-randomized study. Preoperatively and 8 weeks postoperatively, a comprehensive set of assessments was performed. This included optical biometry (IOLMaster 700; Carl Zeiss Meditec AG; Germany), high-resolution OCT imaging utilizing both the CASIA II (Tomey; Japan) and the MS-39 (CSO; Italy), and Scheimpflug imaging (Oculus; Germany). Additionally, autorefraction and subjective refraction measurements were conducted at each time point. Visual acuity assessments were also carried out for distance, intermediate, and near vision evaluated only at the 8-week visit.

Results

100 eyes of 50 patients were included in the study. The mean age of the cohort was 72.28 years (SD: 1.11). At the time of analysis, 48 datasets, including the final post-operative visit, were available. Right eyes showed a mean axial length of 23.12 mm (0.14) and a mean post-operative anterior chamber depth of 4.67 mm (0.05), while left eyes showed 23.05 mm (0.14) and 4.61 mm (0.05), respectively. The mean uncorrected distance visual acuity of both eyes was 0.01 (0.02), while the mean best-corrected distance visual acuity was -0.11 (0.02). The mean uncorrected intermediate visual acuity at a distance of 66 cm for both eyes was 0.23 (0.02), with best correction being 0.18 (0.02). Mean absolute refractive error was 0.21 D.

Conclusions

The Clareon IOL is safe and user-friendly. The post-operative anterior chamber depth corresponds to the expected values, indicating high stability within the capsular bag. The post-operative refractive outcome is excellent without any refractive surprise.

The AI-based prediction of post-operative ACD will be conducted once the two missing final visits have occurred and will be presented at the ESCRS in Barcelona.