Comparison of the diagnostic performance of two spectral-domain OCT machines in detecting glaucoma in high and non-high axial myopia eyes.

PURPOSE: To compare the diagnostic performance of two spectral-domain optical coherence tomography (SD-OCT) systems, Spectralis and Cirrus, in detecting glaucoma among eyes with and without high axial myopia, using peripapillary retinal nerve fiber layer (pRNFL), macular full-thickness, and ganglion cell-inner plexiform layer (GCIPL) measurements.

METHODS

This prospective cross-sectional study included 333 eyes from 333 adults at a tertiary referral center. Participants were classified into four groups according to axial length (high myopia was defined as axial length ≥ 26.0 mm) and glaucoma status: non-highly myopic normal (n = 130), non-highly myopic glaucomatous (n = 101), highly myopic normal (n = 20), and highly myopic glaucomatous (n = 82). Thickness parameters, including pRNFL, macular full-thickness, and GCIPL, from both SD-OCT systems were analyzed, and diagnostic performance was assessed using receiver operating characteristic curves and area under the curve (AUC) comparisons.

RESULTS

Both devices detected significant thinning across pRNFL, macular full-thickness, and GCIPL measurements in glaucomatous eyes. In non-highly myopic eyes, diagnostic performance was generally comparable, although Spectralis demonstrated superior performance in the nasal pRNFL quadrant (P < 0.01). In highly myopic eyes, Spectralis outperformed Cirrus in several macular full-thickness subfields (e.g., inner and outer temporal, inner inferior, and inner superior), whereas Cirrus exhibited better performance in superior pRNFL. GCIPL parameters performed equally well in both devices, with AUC values exceeding 0.89 in highly myopic eyes.

CONCLUSION

Both SD-OCT systems demonstrated reliable diagnostic abilities in eyes with and without high axial myopia. Spectralis showed advantages in specific macular full-thickness parameters in high axial myopia, and GCIPL metrics remained consistent and robust across both devices.