Purchase this article with an account.
M. Monhart, E. Buerki, H. Bebie, A. Palmowski–Wolfe; Calculation of the Abnormal Response Area as an Indicator of Visual Field Changes . Invest. Ophthalmol. Vis. Sci. 2006;47(13):3982.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The study investigates the hypothesis that characteristic deviations of a subjects individual cumulative defect curve (a ranking of all local deviations) compared to the 50% percentile normal defect curve can be used to identify early glaucomatous changes not classified as abnormal based on standard global visual field indices as MD and sLV (PSD).
The area between a normal cumulative defect curve shifted to an individual level of diffuse defect and the respective individual defect curve is calculated and designated ARA for Abnormal Response Area. In 3 different normal populations that underwent white/white (SAP) and blue/yellow (SWAP) perimetry, the 95% confidence limit for ARA is calculated. 368 visual field examinations of Glaucoma Suspects based on abnormal results with the HRT–II examination were evaluated. All visual field tests were taken using the Octopus 101 program G2. 282 were white/white examinations, 86 were blue/yellow examinations, all of them with global indices in the normal range.
The confidence intervals for the ARA index varied largely according to the strictness of the normal value group. The white/white 95% confidence limit for the 3 normal groups was 53.3, 95.9 and 101 respectively. The blue/yellow 95% confidence limit was 56.5, 123.2 and 143.5 respectively. Depending on the normality group chosen the ARA showed between 0..31% of abnormal results for visual fields with normal global indices. The higher percentage results originated from blue/yellow perimetry.
The mathematical identification of the Abnormal Response Area suggests potential in recognizing early defects in visual fields that would be considered normal when applying conventional evaluation methods. Sufficiently large and thoroughly examined normal populations are required to calculate usable confidence intervals. This provided, ARA may be a promising method for evaluating psychophysical tests, adding information to the traditional global indices MD and sLV (PSD).
This PDF is available to Subscribers Only