Posterior Eye Curvature as a Biomarker for Differentiating Pathologic Myopia From High Myopia

Abstract

Purpose: To characterize posterior eye curvature and to assess its diagnostic performance in differentiating pathologic myopia (PM) from simple high myopia (SHM).

Design: Population-based, cross-sectional study.

Participants: A total of 790 eyes from 790 participants (mean age: 60.6 ± 8.3 years; 59% female) were randomly selected from the Beijing Eye Study, including 406 nonmyopic eyes, 175 eyes with mild myopia, 102 eyes with moderate myopia, 76 eyes with simple SHM, and 31 eyes with PM.

Methods: Posterior eye curvature was calculated using the outer boundary of the retinal pigment epithelium layer, derived from automatically segmented spectral-domain optical coherence tomography images. Three curvature parameters were computed: curvature^mean, curvature^max and curvature^macula.

Main outcome measures: The values and spatial distribution of curvature parameters across different myopia groups, and their diagnostic performance in distinguishing PM from SHM.

Results: Higher curvature parameters were significantly associated with older age, longer axial length (AL), more myopic refractive error (RE), lower best-corrected visual activity, and smaller subfoveal choroidal thickness (all P < .001). These parameters increased progressively with myopia severity (all P < .001). Unlike traditional metrics such as AL, which demonstrated a steady increase across the entire spectrum of myopia, curvature parameters showed a significantly sharper increase between SHM and PM, compared to other pairwise comparisons (nonmyopia vs mild myopia, mild vs moderate myopia, and moderate myopia vs SHM) (all P < .001). After adjusting for age, AL, and RE, the curvature parameters remained significantly associated with the occurrence of PM (all P ≤ .003). In distinguishing PM from SHM, curvature^max achieved the highest diagnostic performance, with an area under the receiver operating characteristic curve (AUROC) of 0.92 (95% CI, 0.87-0.98), followed by 0.86 (95% CI, 0.78-0.94) for curvature^mean, 0.84 (95% CI, 0.76-0.93) for curvature^macula, 0.75 (95% CI, 0.62-0.89) for AL, and 0.76 (95% CI, 0.66-0.86) for RE. The performance of curvature^max was significantly higher than AL (P = .020) and RE (P = .006).

Conclusions: Posterior eye curvature increased with myopia severity and outperformed traditional AL and RE, suggesting its potential as a desirable biomarker in differentiating PM from SHM. Further research, particularly longitudinal studies, is warranted to evaluate whether the curvature can predict myopia prognosis and the development of pathologic changes.