Optic disc morphology and myopia severity in eyes with peripapillary hyperreflective ovoid mass-like structures

To compare optic disc morphology, visual field parameters, and OCT measurements between myopic eyes with and without peripapillary hyperreflective ovoid mass-like structures (PHOMS), and to identify factors associated with PHOMS.

This retrospective case-control study included 35 myopic patients (60 eyes) with PHOMS and 23 myopic patients (46 eyes) without PHOMS. All participants underwent comprehensive ophthalmic examinations including refraction, visual field testing, fundus photography, and enhanced depth imaging OCT (EDI-OCT). PHOMS height, peripapillary RNFL thickness, and BMO-MRW were measured. Propensity score matching (1:1 on age, sex, and spherical equivalent) was performed, and generalized estimating equations (GEE) accounted for inter-eye correlation. False discovery rate (FDR) correction (q = 0.05) was applied for univariable and correlation analyses; no multiplicity adjustment was made for the multivariable logistic regression.

After matching (39 eyes per group), the PHOMS group exhibited a lower optic disc ovality index (ODOI) (0.746 ± 0.072 vs. 0.803 ± 0.072, P = 0.006) and a smaller cup-to-disc ratio (CDR) (P = 0.001). Multivariable GEE logistic regression identified ODOI as the only independent factor significantly associated with the presence of PHOMS (OR = 3.332, 95% CI: 1.548-7.172, P = 0.002), whereas spherical equivalent (SE) was not an independent factor after adjustment (OR = 1.277, 95% CI: 0.826-1.975, P = 0.270). However, a combined predictive model incorporating SE, ODOI, and CDR achieved the highest diagnostic performance with an area under the curve (AUC) of 0.830 (95% CI: 0.746-0.913). Under the fully adjusted GEE model, the PHOMS group showed significantly thinner nasal peripapillary retinal nerve fiber layer (pRNFL) (P = 0.009), thicker temporal and inferior pRNFL (P = 0.003 and P = 0.030, respectively), and a larger Bruch's membrane opening-minimum rim width (BMO-MRW) (P < 0.001). Physiological blind spot enlargement was more frequent in the PHOMS group (P = 0.031), whereas global visual field indices showed no significant differences. PHOMS height correlated positively with pseudopapilledema grade (rho = 0.703, P < 0.001) and average pRNFL thickness (rho = 0.480, P < 0.001).

PHOMS were associated with a lower optic disc ovality index (ODOI, indicating greater optic disc tilt) and a smaller CDR. Although myopia severity (SE) was higher in PHOMS eyes in unadjusted comparisons, only ODOI remained independently associated with PHOMS in multivariable analysis. A combined prediction model using SE, ODOI, and CDR predicted PHOMS presence with good accuracy (AUC = 0.830). Eyes with PHOMS exhibit a characteristic topography of nasal pRNFL thinning and temporal/inferior thickening without evidence of glaucomatous damage. These findings support an association between PHOMS and myopia-related disc remodeling but do not prove that PHOMS directly cause optic disc damage.