Choroidal thickness as a biomarker for axial length change

Matilda Biba,Gareth Lingham,Emmanuel Kobia Acquah,James Loughman,Daniel Flitcroft

DOI: https://doi.org/10.1111/aos.16119

2024-01-01

Acta Ophthalmologica

Abstract:Aims/Purpose: Assessing axial eye growth is a key component in myopia management; however, access to biometry is limited in primary eye care settings. This study evaluated choroidal thickness, compared to spherical equivalent refraction, as a potential biomarker of axial elongation. Methods: The Myopia Outcome Study of Atropine in Children (MOSAIC) is a double‐blind, randomized controlled trial of 0.01% atropine eye drops. Irish myopic participants aged 6–16 years used either 0.01% atropine or placebo eye drops nightly for 2 years. Spherical equivalent refraction (SER) was measured by cycloplegic autorefraction, axial length (AL) by partial coherence interferometry and choroidal thickness (ChT) by Swept Source‐OCT. The respective 12‐month changes in both eyes were included in mixed linear regression analysis, while controlling for treatment group and within‐subject correlation. ROC analysis was conducted to assess the ability of ChT and/or SER to identify eyes in the top 25% (fast progressors) of AL change. Results: Of the 250 participants enrolled in MOSAIC (mean age 11.8 years, 62% female), 217 (87%) completed the 12‐month visit. Mean 12‐month change in AL, ChT and SER were 0.268 ± 0.19 mm, −3.62 ± 33.3 um and −0.437 ± 0.346 DS respectively. Analysis revealed a significant association between change in ChT and AL (−0.01 ± 0.002 mm per 10 μm; p < 0.001) and change in SER and AL (−0.28 ± 0.02 mm per 1D; p < 0.001). When both SER and ChT were combined in a single model, a 1D myopic change in SER and a 10um decrease in ChT were associated with a 0.27 mm ( p < 0.001) and 0.007 mm increase in AL ( p = 0.002), respectively. ROC analysis revealed that change in ChT was a stronger predictor of fast AL progression (AUC = 0.77) compared to SER change (AUC = 0.72), with the combined model performing better than either variable alone (AUC = 0.80). Conclusions: ChT change was significantly associated with AL change over 12 months and better than SER alone at identifying fast AL progressors, with a combined SER and ChT model achieving the best results.

ophthalmology