Purchase this article with an account.
J. Caprioli, K. Nouri–Mahdave, D. Hoffman, D. Gaasterland; Predicting Probability of Visual Field Progression . Invest. Ophthalmol. Vis. Sci. 2004;45(13):3469.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose:To predict the probability of glaucomatous visual field (VF) progression with clinical and perimetric data. Methods:Two hundred thirteen eyes (of 170 patients) enrolled in Advanced Glaucoma Intervention Study (AGIS) who had ≥8 years of follow–up, a baseline VF score ≤16, and a good reliability score were selected. Progression at 8 years was determined with pointwise linear regression analysis (PLR) using a two–omitting algorithm. The course of VF over the first 4 years of follow–up was summarized by calculating an index, "sum of slopes", consisting of the sum of all slopes of VF thresholds demonstrating a p value of < 0.05 when PLR was performed on 4 year data. The following parameters were entered in a logistic regression model to predict 8 year outcomes from 4 year data: sum of slopes at 4 years, age, IOP fluctuation and mean IOP during the first 4 years of follow–up, and AGIS VF score at 4 years. Results:Seventy–six eyes (36%) progressed after 8 years of follow–up according to PLR. Three parameters were predictive of subsequent VF progression, as identified at 8 years (predictive power: 77.5%): more negative sum of slopes (i.e. faster or more extensive deterioration) during first 4 years (p < 0.001), higher mean IOP during first 4 years (p = 0.019), and older age (p = 0.046). Removing mean IOP and age from the regression model did not decrease the predictive power (77.0%). Conclusions:Visual field "sum of slopes" during early follow–up was the strongest predictor of future course of glaucoma. Sum of slopes proved to be an easily calculated index that can be used to estimate the probability of subsequent VF worsening. This probability can be combined with other clinical information for more effective clinical predictions and decision–making.
This PDF is available to Subscribers Only