Purpose
Current optic disc margin (DM)-based neuroretinal rim assessment lacks a solid anatomic and geometric basis. Spectral domain OCT (SD-OCT)-derived Bruch’s membrane opening minimum rim width (BMO-MRW) has higher diagnostic value than DM-based measures and relates better to the visual field. We characterised BMO-MRW in a normal Caucasian population and modelled its variability on a range of ocular and demographic parameters.
Methods
Subjects aged between 20 and 90 years (approximately same number in each decade group) were enrolled in 5 centres. Inclusion criteria were: (1) normal ocular and visual field exam; (2) visual acuity ≥ 6/12 and (3) refraction within ±6 D and 2 D astigmatism. SD-OCT images of the optic nerve head (24 radial scans centred on BMO, Spectralis), acquired relative to the eye-specific fovea to BMO centre (FoBMO) axis, and ocular biometry data were obtained. Variation of BMO-MRW was analysed with respect to age, sectors relative to FoBMO axis, and BMO size and shape. BMO-MRW data were compared to an independent set of DM-based scanning laser tomography rim area (DM-RA) measurements from the Moorfields Regression Analysis database.
Results
There were 246 eyes of 246 subjects in the study. The FoBMO angle ranged +2° to -18° (median, -7°), relative to horizontal, and was not associated with axial length or refractive error. BMO area (median, 1.74 mm2) was not associated with age or axial length. The long axis orientation of BMO was predominantly vertical (with 61% of the observations between 5 and 7 ‘o’ clock relative to the FoBMO axis) and unrelated to FoBMO angle. Mean BMO-MRW declined significantly with age (R2=0.15), in sharp contrast with DM-RA (R2=0.01, Fig. 1). The steepest decline, measured in either absolute units or adjusted for the population sector mean BMO-MRW, occurred inferiorly (Fig. 2). In multivariate analyses, BMO shape, BMO area and age significantly influenced BMO-MRW explaining ~15% of its variation.
Conclusions
(1) BMO-MRW has a stronger relationship with age than DM-RA, likely indicating a more accurate marker of age-related retinal ganglion cell loss; (2) the sectoral patterns strongly mimic the sectorally related patterns of longitudinal age-related change in normals and progressive glaucoma published previously; (3) these new normative data will allow more accurate assessment of the optic nerve head for glaucoma.
Keywords: 550 imaging/image analysis: clinical •
629 optic nerve •
552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)