Macular SD-OCT Outcome Measures in Glaucoma: Comparison of Local Structure-Function Relationships and Dynamic Range

Abstract

Purpose : There is scant evidence regarding comparative utility of various macular SD-OCT thickness parameters for detection of glaucoma progression. We have previously found the local variability for all macular parameters to be very low and uniform across the macula. The goal of this study is to compare the performance of inner macular thickness parameters with regard to local structure-function (SF) relationships and dynamic range of measurements.

Methods : One hundred fifteen glaucomatous eyes from 102 patients, including 17 eyes with intact achromatic visual field, and 16 normal eyes (8 subjects) with macular SD-OCT images (Spectralis) and 10-2 visual fields were enrolled. Macular images were segmented and inner plexiform layer (IPL), ganglion cell layer (GCL), and GC/IPL thicknesses were calculated for an 8x8 array of 3-degree superpixels. The main outcome measures used to compare macular parameters were: 1) local structure-function (SF) relationships within the central 18 degrees of the macula after adjusting for macular retinal ganglion cell displacement, and 2) dynamic range of measurements, defined as the difference between the median of the top highest 5% of measurements and the bottom plateau of OCT measurements (the intercept of Hood’s simple linear model) and the change point, i.e. the local total deviation value where the macular paramters reach the bottom plateau.

Results : The average visual field mean deviation was −6.8 (± 5.4) dB in the glaucoma group. The strength of SF relationships was similar among all thickness measures (rho = 0.571, 0.519, and 0.578 for GCL, IPL, and GC/IPL, respectively; p >0.05 for the difference). The highest SF correlations coincided with the peak of the GCL thickness (rho = 0.610, 0.618, and 0.637). The dynamic range was highest for the GC/IPL (57μ) followed by GCL (36μ), and IPL (22μ)(Figure). The change points were very similar among all the parameters as follows: IPL=−8.4 dB, GCL=−8.7 dB, and, GC/IPL=−8.6 dB.

Conclusions : The strength of structure-function relationships and the latter's point of change were similar for all inner macular parameters. Given the similar variability for the 3 macular parameters, the higher dynamic range, and the less demanding segmentation, GC/IPL provides the best compromise for detection of change over time with current SD-OCT technology.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.