Purchase this article with an account.
Jack Phu, Michael Kalloniatis; Static automated perimetry using a new head-mounted virtual reality platform, Virtual Field, compared with the Humphrey Field Analyzer in glaucoma and optic nerve disease. Invest. Ophthalmol. Vis. Sci. 2021;62(8):3364.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To compare the visual field outputs obtained using Virtual Field against the Humphrey Field Analyzer (HFA) in subjects with glaucoma or optic nerve head disease, and to determine its potential deployment as a portable, low-cost alternative for perimetric testing.
One eye (the eye with more severe disease) from 43 subjects with open angle glaucoma or optic nerve disease underwent testing using both Virtual Field and the HFA. Both devices presented Goldmann size III stimuli in the 24-2 test grid arrangement. Virtual Field uses a screen with a background luminance of 0.218 cd.m-2, whereas the HFA uses a projection system with a background luminance of 10 cd.m-2. All patients undertook three tests on each device in random order. We compared conventional global indices (mean deviation [MD], pattern standard deviation [PSD], "cluster" criterion pass/fail, Glaucoma Hemifield Test [GHT]), pointwise sensitivity and probability scores, and the overall number of defects found using each device, after averaging all indices across all three tests.
MD and PSD were correlated between both devices (Fig 1). There were differences in the individual decibel values reported by each device due to the differences in background luminance and hence retinal adaptation. Bland-Altman analysis revealed a bias of 6.1 dB, with 95% limits of agreement between -1.8 to 14.0 dB, with the HFA results returning higher readings. 10 (23%) subjects showed more defects on the HFA, 3 (7%) had more defects on Virtual Field, and the others had no differences between devices. On average, the HFA identified 3 more defects on the pattern deviation map compared to Virtual Field. Using a combinatory criterion of PSD (>2dB), GHT (outside normal limits) and "cluster" cut-offs, there were no significant differences between devices in identifying the presence of disease (6 cases identified by only Virtual Field, 1 only by the HFA, 31 by both, and 5 by neither).
Virtual Field is a head-mounted perimeter that produces functionally indistinguishable results to the HFA, offering a low-cost, portable alternative to specialised clinic-based perimetry devices.
This is a 2021 ARVO Annual Meeting abstract.
Comparison of mean deviation (dB) and pattern pattern standard (dB) results obtained using HFA (y-axis) versus Virtual Field (x-axis), with linear regression results shown in the insets.
This PDF is available to Subscribers Only