Diagnostic Innovation in Glaucoma Study (DIGS): Investigation of Floor Effects for Optical Coherence Tomography Structural Measurements in Advanced Glaucoma

Abstract

Purpose : The “floor effect” in optical imaging is defined as the tissue thickness remaining after which no further structural loss can be detected while functional changes (due to glaucomatous progression or aging) are still detectable. This study describes the relative floor effects for minimum rim width (MRW), peripapillary retinal nerve fiber layer thickness (RNFL) and retinal ganglion cell – inner plexiform layer thickness (GCIPL) measured with spectral domain optical coherence tomography (SDOCT).

Methods : 87 eyes of 59 patients with advanced glaucoma (SAP MD ≤ -12 dB) were included in the study group. A stable group of 41 eyes of 27 glaucoma patients with moderate to advanced glaucoma (MD ≤ -8 dB) with repeated tests over 5 weeks was used to estimate the floor effect. All eyes underwent two 24-2 SAP exams, Spectralis SDOCT circular scans, ONH volume scans, ONH radial scans and macular volume scans within 2 years. The RNFL, MRW and GCIPL layer thicknesses were measured using the San Diego Automated Layer Segmentation Algorithm (SALSA). The floor effect was defined as the areas of the structural measurements with a loss smaller than the 1^st percent variability of the stable group.

Results : Mean age and MD of the advanced glaucoma group were 73 years (range 61, 88) and -17 dB (-12.5, -27). Mean age and MD of the stable group were 70 years (range 63, 86) and -10.4 dB (-8.3, -23). The average (± S.D.) floor effects were 38 µm (± 4.2) for circular RNFL thickness, 36 µm (± 5.9) for volume RNFL thickness, 105 µm (± 15.9) for MRW and 38 µm (± 3.4) for the GCIPL. The mean percentage of image area that did not reach the floor were 14% for circular RNFL thickness, 25% for volume RNFL thickness, 19% for the MRW and 36 % for GCIPL. When eyes with only local defects (floor RNFL thickness from volume scan < 50%) were excluded (31 eyes), the mean percentage of image area that did not reach the floor were 5% for circular RNFL thickness, 11% for volume RNFL thickness, 8% for the MRW and 22 % for GCIPL.

Conclusions : Different SD-OCT structural measurements display different floor effects in advanced glaucoma. Areas that are less damaged may still be used to monitor glaucoma. Longitudinal studies are needed to determine if relative floor effects among the parameters investigated reflect the dynamic range of measurements useful for detecting change over time.

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