Clinicians' Use of Quantitative Information when Assessing Structural Progression in Glaucoma

Abstract

Purpose : OCT scans contain large amounts of information which clinicians must consider when assessing the rate of glaucomatous progression. Yet images of the scans are small, so clinicians rely heavily on quantitative information provided in the form of global and sectoral layer thicknesses. We investigated which of these quantitative measures are most closely related to the subjective assessment of glaucoma experts who had all the OCT information available.

Methods : Eleven glaucoma specialists independently assigned scores for the rate of structural progression based on series of 5 biannual Heidelberg Spectralis OCT printouts from 100 glaucoma or glaucoma suspect eyes, from 51 participants in the Portland Progression Project longitudinal study. They also reviewed 20 of the series twice to assess repeatability. Scores were on a scale from 1 (improvement) to 7 (very rapid progression), and were averaged among clinicians. Generalized estimating equation linear models, weighted by intra-eye inter-clinician score variability, were used to predict the mean clinician score from the rates of change of Retinal Nerve Fiber Layer Thickness (RNFLT) or Bruch’s Membrane Opening Minimum Rim Width (MRW), either globally or in the most rapidly thinning of the six sectors presented by the instrument’s software.

Results : The average global RNFLT within the series was 79.3µm (range 41.4 to 126.6); average global MRW was 214.4µm. Of individual clinician scores, 95% varied by ≤1 point when repeated. The average mean clinician score was 2.6 (standard deviation 0.7, range 1.5 to 4.8). This score was more strongly correlated with the rate of change of RNFLT in the most rapidly changing sector (R²=0.657 when expressed in %/y, or R²=0.582 using µm/y) than with the global RNFLT rate (R²=0.372 and 0.305 respectively); see Figure 1. These comparisons persisted, albeit with smaller differences in R², when the three outliers with apparent thickening of the RNFLT were excluded. Correlations with the rate of MRW thinning were consistently weaker (maximum R²=0.149 for the most rapidly changing sector in %/y) than those for RNFLT for both global and sectoral analyses; see Figure 2.

Conclusions : The percentage rate of change of RNFLT in the most rapidly changing sector predicted experts’ assessment of the rate of structural progression better than global rates. Sectoral RNFLT rates may be a useful addition to current clinical printouts.

This is a 2021 ARVO Annual Meeting abstract.