The Impact of Normal Aging and Definitions of Progression on Detecting Retinal Nerve Fiber Layer Changes in Optical Coherence Tomography Imaging

Abstract

Purpose : When attempting to identify progressive change in retinal nerve fiber layer (RNFL) thickness on optical coherence tomography (OCT) imaging (such as due to glaucoma), age-related thinning may account for a large proportion of changes assumed to be pathological. This may result in high false positive rates of detecting progression, which could lead to inaccurate clinical management decisions. In this study, we evaluated the rates of false detection of progression in normal eyes using current and proposed methods to account for age-related RNFL thickness changes.

Methods : A total of 77 eyes of 46 normal participants seen at 5.9 ± 2.7 visits over 3.4 ± 1.9 years were included to obtain the mean and 5% lower limit of the range of normal age-related RNFL thickness change over time. Variability and age-related change estimates were then obtained and used to recreate 10,000 simulated series of “real-world” RNFL measurements, which were used to compare the false positive rates of different definitions of progression.

Results : The mean rate of RNFL thickness change over time was -0.50 ± 0.25 μm/year (P < 0.001). Almost 25% of simulated eyes were identified as having progressed with a slope significantly less than 0 at some point during the first 5 years of follow-up with annual testing. Additionally requiring the slope to be more negative than the 5% lower limit of normal aging still identified 18% of eyes as having progressed. In contrast, definitions that tested the significance of the slope relative to the mean or 5% lower limit of aging (also requiring a slope exceeding the 5% lower limit) reduced false-positive rates to 8% and 4% respectively.

Conclusions : False detection of progression rates when using OCT imaging can be substantially reduced by testing the significance of its longitudinal change relative to normal age-related changes. This approach, which could be described as a “mean trend deviation” analysis, can help clinicians differentiate between age-related and glaucomatous change.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.