- PMID: 41057223
- DOI: 10.1136/bjo-2025-328223
Abstract
Aims: To compare the predictive efficacy of precataract extraction fundus photography, optical coherence tomography (OCT) and multifocal electroretinogram (mfERG) in determining 3-month postcataract extraction best-corrected visual acuity (BCVA) in patients with high myopia complicated by cataracts.
Methods: This retrospective cohort study included 58 highly myopic eyes that underwent cataract surgery. Preoperative assessments included macular lesion grading from fundus photography based on the META-pathologic myopia classification; assessment of outer retinal structural integrity, specifically ellipsoid zone (EZ) status, via OCT and mfERG parameters, including P1 response density and implicit time from five concentric rings. Postoperative BCVA was recorded at 3 months. Statistical analyses included Pearson correlation, multiple linear regression and receiver operating characteristic curve analysis to evaluate the relationship and predictive power of each modality and a combined model.
Results: The study included 58 eyes with a mean age of 62.7±10.0 years. Preoperative BCVA, myopic maculopathy grading, EZ status and central mfERG parameters all showed strong correlations with postoperative BCVA (all p<0.001). In the multivariate linear regression model, preoperative BCVA (β=0.270, p<0.001), myopic maculopathy grading (β=0.075, p=0.029), EZ status (β=0.093, p=0.032) and central mfERG P1 amplitude density (β=-0.003, p=0.013) were all significant independent predictors of postoperative BCVA. An OCT-based model demonstrated the highest area under the curve (AUC) among single modalities (AUC=0.875). A combined mixed model incorporating all variables achieved the highest predictive accuracy (AUC=0.934).
Conclusion: Preoperative assessments of fundus morphology, OCT-derived microstructures and retinal function are all valuable, independent predictors of visual outcomes after cataract surgery in highly myopic eyes.
Keywords: Lens and zonules; Retina.
© Author(s) (or their employer(s)) 2026. No commercial re-use. See rights and permissions. Published by BMJ Group.
Conflict of interest statement
Competing interests: None declared.

