Purchase this article with an account.
Dario Romano, Giovanni Montesano, Paolo Fogagnolo, Allison M McKendrick, Andrew Turpin, Francesco Oddone, Paolo Lanzetta, Paolo Brusini, Andrea Perdicchi, Chris A Johnson, David F Garway-Heath, David P Crabb, Luca Mario Rossetti; Effect of additional testing locations on the evaluation of macular perimetric defects in glaucoma. Invest. Ophthalmol. Vis. Sci. 2019;60(9):2458. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate if additional macular test locations could better characterize the macular defect in glaucoma.
We analysed data from one eye of 439 healthy subjects and 499 with glaucoma from the validation study of the Compass fundus perimeter (CMP, CenterVue). Subjects were tested with the CMP using a New-Grid containing all 24-2 grid test locations plus 12 additional macular locations (Figure 1). We restricted the analysis to the macula, comparing the macular Mean Deviation (MD) of the New Grid with the traditional 24-2 calculated on the locations within 10 degrees from fixation (Figure 2). We also computed the Defect Detection Rate (DDR) of the two strategies by calculating the percentage of glaucoma subjects that had at least one location outside the 95% normal limits in the macula. This same analysis was conducted separately on four macular sectors (Figure 1). The False Positive Rate (FPR) was calculated as the DDR using the healthy subjects. Confidence intervals for the increase in FPR and DDR with the two grids were calculated via bootstrap.
The DDR increase for the whole macula with New Grid was not significantly larger than the FPR increase. When sectors were analysed, only the increase of the Superior-Nasal DDR (12% [10, 15], Mean [95% CIs]) was significantly larger than the increase in FPR (8% [5, 10], p = 0.0114). The difference in MD between the two grids was significantly negatively correlated with the New Grid MD (p < 0.001), showing a consistent underestimation of the defect with the 24-2 with more advanced damage (Figure 2). Estimated time increase with New Grid was 30 ± 31 seconds for healthy subjects and 39 ± 49 seconds for glaucoma patients.
Additional test locations can significantly change the estimate of the macular damage, but the effect is small, with increased test time. However, they can increase the sensitivity to macular damage detection for the Superior-Nasal sector. This is consistent with previous knowledge on where early macular damage is more likely to appear.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
Figure 1. On the left, New Grid with 24-2 locations in red and additional points in black. The analysed sector is shaded in gray. On the right, bootstrap distributions of FPR and DDR increase of New Grid over the 24-2
Figure 2. Average underestimation of the defect at advanced macular damage with the 24-2
This PDF is available to Subscribers Only