Spectral Domain Optical Coherence Tomography (SDOCT) Optic Nerve Head (ONH) Rim Quantification in Glaucoma and Glaucoma Suspect Eyes Using Anatomic Vs Acquired Image Frame (AIF) Regionalization

Abstract

 

Purpose

 

SDOCT can be used to quantify ONH rim parameters. However, the default SDOCT parameterization is based on AIF orientation, which can vary substantially from the axis of a highly influential anatomic relationship: the vertical offset of the fovea relative to the center of the ONH. The latter can be defined reliably as Bruch’s membrane opening (BMO) centroid. This study compared neuroretinal rim parameters regionalized according to Fovea-to-BMO axis (FoBMO) versus AIF axis.

 

Methods

 

Spectralis SDOCT scans (48, 15° radial B-scans centered on the ONH) were obtained in 190 glaucoma patients and high-risk suspects. Masked technicians delineated the internal limiting membrane (ILM), retinal nerve fiber layer (RNFL) and BMO in half (24) of the B-scans of each eye. The ONH was regionalized into 12 sectors using the FoBMO axis and the AIF axis (Fig. 1). For each sector, two neuroretinal rim parameters were quantified for both regionalization methods: minimum rim width (MRW) defined as the minimum distance from the BMO to the ILM and minimum rim area (MRA) calculated using MRW and its angle above the BMO plane. For both MRW and MRA, differences between FoBMO and AIF axes were calculated.

 

Results

 

Both MRW and MRA were statistically significantly different (Wilcoxon signed-rank test, p<0.05) for 7 of the 12 sectors when FoBMO and AIF regionalizations were compared (Fig. 2). The AIF axis generally tended to result in higher MRW and MRA in Stemp and Tsup sectors and lower MRW and MRA in Tinf and Itemp sectors as compared with FoBMO-based sectors. The largest difference for both MRW and MRA occurred in the Itemp sector where the AIF-based MRW was 12.9 μm thinner and the MRA was 5397 μm2 smaller than the FoBMO-based rim. For individual eyes, these differences can be as large as 90 μm for MRW and 40000 μm2 for MRA, corresponding to 36% and 44% of the average values respectively.

 

Conclusions

 

Because the angle of declination from the ONH center to the fovea influences axon bundle paths, using a FoBMO axis that is common to nearly all human eyes should reduce between subject variability within normative data bases and improve structure/function relation.

 

 

 

Keywords: 627 optic disc • 552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • 550 imaging/image analysis: clinical