Purchase this article with an account.
Navid Amini, Nima Fatehi, Sharon Henry, Joseph Caprioli, Kouros Nouri-Mahdavi; Foveal Slope as A Biomarker for Detection of Glaucoma Progression with Macular OCT Imaging. Invest. Ophthalmol. Vis. Sci. 2018;59(9):2107. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Technical challenges exist in measuring the ganglion cell layer close to the fovea where it matters most for central vision. We tested the hypothesis that the slope of foveal depression can be used as a biomarker for detection of glaucoma progression with macular OCT imaging.
30 glaucoma patients with triplicate images and 67 normal subjects and were enrolled. Macular OCT cubes were segmented to measure individual retinal layers. Surface derivative and thickness functions were used to locate the foveal center. The slope of the rising foveal edge was found along 12 equidistant meridians (Figure 1). Spline curves were fitted to GCIPL inner border to reduce noise and accommodate missing data. Locations demonstrating the highest 25%ile rates of increase in the GCIPL slope were identified. We explored predictors of foveal slope in normal subjects, estimated measurement variability with intraclass correlation coefficients (ICC), built bivariate plots of foveal slopes vs. total deviation (TD) values of paracentral 10-2 locations, and calculated changes of foveal slope in a subset of progressing eyes.
The median (IQR) MD was –7.5 (–1.0 to –22.9) dB in glaucoma patients. None of the demographic or clinical factors, including age and axial length, predicted foveal slopes in normal subjects (p >0.05). The ICCs for foveal slopes ranged from 0.962 (90° meridian) to 0.999 (270° meridian). The median SD for the 12 triplicate slope measurements varied from 0.8 (30° meridian) to 1.3 (90° meridian) μm/mm; therefore, a decrease in foveal slope exceeding 2.3-3.7 μm/mm would represent a significant decline (based on test-retest variability or TRV=√2 x 1.96 x SD). The Spearman correlation coefficient between 4 paracentral location TDs and corresponding foveal slopes was 0.281 (p=0.002; Figure 2). The temporal foveal slope significantly decreased (12.5-52.9 μm/mm) compared to TRV limits in 5 eyes progressing according to VFs 24-30 months after the baseline exam.
Assessment of the foveal slope is promising as a biomarker for detection of glaucoma progression given its high reproducibility, cross-sectional correlation with glaucoma severity, and early findings in progressing eyes.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only