Purchase this article with an account.
Diana Salazar, Esteban Morales, Alessandro Rabiolo, Vicente Capistrano, Mark Lin, Fei Yu, Joseph Caprioli; POINTWISE METHODS TO ASSESS LONG-TERM VISUAL FIELD PROGRESSION IN GLAUCOMA. Invest. Ophthalmol. Vis. Sci. 2019;60(9):2471.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare three methods to evaluate visual field (VF) progression that retain spatial information: Guided Progression Analysis (GPA), pointwise linear regression (PLR), and Glaucoma Rate Index (GRI) and to assess their patterns of progression.
729 eyes of 567 open-angle glaucoma patients with ≥6 reliable VFs and ≥3 years of follow-up were included in this retrospective, observational study. GPA, PLR, GRI, and pointwise rates of change (PRC) were calculated. The proportions of VF series detected as progressing and the times to progression were calculated for each technique. From the entire cohort, a subset of patients with a very high likelihood of progressing was identified with a method known to have high specificity (AGIS score), and the proportions of VF series detected as progressing and times to progression were also compared in this subset. Pairwise agreement between methods was calculated. The status of each test location was determined with all 3 methods: for trend-based methods (PLR, and PRC), the pointwise rates of progression were also measured.
The mean ±SD age of the cohort was 65.6±9.7 years. The mean baseline MD and follow-up were -7.7±5.6 dB and 8.8±2.0 years, respectively. The proportions of eyes labeled as progressing according to GPA, PLR and GRI criteria were 27.7%, 33.5%, and 52.9%, respectively (all pairwise comparisons p<0.001). Moderate agreement of progression was seen between all methods (GRI/PRC k=0.59, PLR/GPA k=0.56 and GRI/GPA k=0.43). GRI had a much faster time to detect progression with a median of 4.3 years compared to GPA (14.5 years) and PLR (14.7 years) (p<0.001). In the subset with a high likelihood of progression, the proportion of detection and time to detect progression for GPA, PLR, and GRI were 74% and 6.0y, 81%, and 4.7y and 93% and 2.5y, respectively. A significant difference was seen between GRI and the other groups for both measurements while no significant differences were seen between GPA and PLR (Figure1). The proportion of locations decaying was higher for PRC (23%) followed by PLR (9%) and GPA (6%).
Pointwise methods to detect glaucoma progression did not agree well regarding either the global VF or pointwise series. GRI had the highest detection of progression and was the fastest to detect progression in both the entire cohort and the subset with a very high likelihood of progression.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only