gms | German Medical Science

97. Versammlung des Vereins Rhein-Mainischer Augenärzte

Verein Rhein-Mainischer Augenärzte

09.11.2024, Mainz

KI meets Vabysmo

Meeting Abstract

  • Anna Theresa Lorenz - Sulzbach/Saar
  • A. M. Messias - Sulzbach/Saar
  • W. Darwisch - Sulzbach/Saar
  • P. K. Roberts - Sulzbach/Saar
  • K. T. Boden - Sulzbach/Saar
  • P. Szurman - Sulzbach/Saar
  • B. V. Stanzel - Sulzbach/Saar

Verein Rhein-Mainischer Augenärzte. 97.Versammlung des Vereins Rhein-Mainischer Augenärzte. Mainz, 09.-09.11.2024. Düsseldorf: German Medical Science GMS Publishing House; 2025. Doc24rma05

doi: 10.3205/24rma05, urn:nbn:de:0183-24rma059

Published: May 16, 2025

© 2025 Lorenz et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Purpose: To assess the impact of faricimab treatment on neovascular age-related macular degeneration (nAMD) and diabetic macular edema (DME) using an artificial intelligence-based method for quantifying retinal fluid volume on spectral-domain optical coherence tomography (SD-OCT).

Methods: The study retrospectively included patients with nAMD and DME; either treatment-naïve, or previously on another intravitreal medication (switchers), who underwent 4 monthly intravitreal faricimab injections. SD-OCT (Heidelberg Engineering) was captured at baseline and 16 weeks, then processed using the Fluid Monitor® (RetInSight, Vienna, Austria). Fluid volumes in the central retinal 1 mm were quantified and stratified into 3 compartments: pigment epithelial detachment, subretinal fluid, and intraretinal fluid. The sum of these compartments was computed (SF), and their relationship with central subfield thickness (CST) was explored using standard linear regression analysis. Corrected distance visual acuity (CDVA) data were extracted from medical records.

Results: Thirty-four nAMD (25 switchers) and 21 DME (20 switchers) eyes were included. The mean SF (nL) was 126.68±17.24 and 37.84±8.31 at baseline, significantly reducing to 80.78±15.56 (p<0.0001) and 15.28±4.94 (p<0.0001) for nAMD and DME, respectively. Mean CST (µm) also significantly reduced from 405.12±24.95 and 354.97±15.89 to 320.33±19.80 (p=0.0001) and 302.41±11.55 (p<0.0001) in nAMD and DME, respectively. The mean intraindividual change observed between baseline and 16 weeks was larger using SF than with CST for nAMD (36.5% and 17.6%, respectively) and for DME (56.2% and 13.1%, respectively). A similar pattern was observed for each retinal compartment. At baseline and postoperatively, the coefficient of determination (R-squared) between SF and CST was 0.799 (p<0.0001) and 0.792 (p<0.0001) for nAMD and 0.511 (p=0.0056) and 0.453 (p=0.004) for DME. CDVA (logMAR) remained stable both in the nAMD group (0.52±0.07 to 0.46±0.06; p=0.1452), and in the DME group (0.48±0.08 to 0.44±0.07; p=0.2694).

Conclusions: When switched to and/or uploaded with faricimab, total fluid decreased by 37% in nAMD and 56% in DME. The SF reduction was roughly 2 and 4-fold larger than the CST reduction for nAMD and DME, respectively. Fluid volume appears to offer more granular insights into retinal leakage and thus warrants further validation in larger real-life cohorts.