Purchase this article with an account.
Steven L Mansberger, Robert Kinast, C Gustavo De Moraes, Donald L Budenz, Jin Wook Jeoung, John Lind, Jonathan S. Myers, Kouros Nouri-Mahdavi, Lindsay Anne Rhodes, Nicholas G Strouthidis, Teresa C Chen, Stuart Keith Gardiner; Clinicians’ Use of Global Indices when Assessing Functional Progression in Glaucoma. Invest. Ophthalmol. Vis. Sci. 2021;62(8):1842.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Clinicians integrate a wealth of information when assessing the rate of glaucomatous functional progression and they use instrument-derived visual field printouts and global indices to assist this task. We sought to determine the global indices that best correlate with the subjective assessment of the rate of visual field progression by glaucoma experts.
Expert glaucoma specialists (n=11) reviewed a series of visual field printouts for 5 biannual visits from 100 glaucoma or glaucoma suspect eyes of 51 participants enrolled in the ongoing Portland Progression Project. Each clinician independently scored each series from 1 (improvement) to 7 (very rapid progression). They also scored 20 of those eyes twice to assess intra-clinician repeatability. Generalized estimating equation models were used to predict the average score among the clinicians from the rates of change of each of the available global indices; Mean Deviation (MD), Pattern Standard Deviation (PSD) and Visual Field Index (VFI).
The average MD of study eyes was -2.4dB (range -16.8 to +2.8dB); 18% had average MD < -6dB. When assessing intra-clinician repeatability, 94% of individual clinician scores varied by ≤1 point. The mean inter-clinician score was also highly repeatable, with an intraclass correlation coefficient of 0.95. This score was more strongly correlated with the rate of change of VFI (r=-0.85) than either MD (r=-0.80) or PSD (r=0.64); see Figure 1. In bivariable regression models using two global indices, the rate of VFI change was the sole significant predictor. However, among eyes with average VFI>99% (n=25) for which VFI is near its maximum possible value, these correlations were r=-0.40 for MD, r=-0.53 for VFI, and r=0.59 for PSD (Figure 2); and the rate of PSD was the strongest predictor in regression models.
Overall, the mean assessment of functional change from a group of highly experienced glaucoma clinicians correlated better with rates of change of VFI than MD or PSD. However, in early glaucoma or glaucoma suspect eyes, clinician judgment correlated slightly better with changes in PSD, than MD or VFI. This may be related to clinicians identifying localized changes at the earliest stages of the disease. Future studies can use this information to better monitor the rate of progression, and use this data to develop improved quantifications of the rate of functional progression.
This is a 2021 ARVO Annual Meeting abstract.
This PDF is available to Subscribers Only