Purchase this article with an account.
Yu Xiang George Kong, Mingguang He, Jonathan Crowston, Keith R Martin, Algis J Vingrys; Longitudinal comparison of visual field outcomes obtained by a tablet perimeter and those returned by Humphrey Field Analyzer. Invest. Ophthalmol. Vis. Sci. 2017;58(8):2866.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Performing perimetry using modern tablet devices enables portable and more frequent perimetry testing. We aim to establish the medium term repeatability of perimetric results from an iPad-based app (Melbourne Rapid Fields, MRF) compared to Humphrey Field Analyzer (HFA) 24-2 SITA-Standard.
Thirty-seven patients with either treated glaucoma or ocular hypertension (n=32) or glaucoma suspects (n=5) were recruited into a 6-month longitudinal clinical study with visits at baseline, 2, 4 and 6 months. At each visit patients undertook visual field assessment using the HFA, the MRF radial pattern (MRF-r) and the MRF 24-2 modified grid pattern (MRF-g). Exclusion criteria were: inability to understand English instructions given by the iPad, acuity worse than 6/12, intraocular surgery in past 6 months and poor reliability indices on the HFA at baseline (>15% false positives, >20% false negatives or Fixation loss). Comparisons were made by establishing correlations (Pearson) and using Bland-Altman methods (Bias, 95% Limits of Agreement, LoA) for the summary statistics (Mean Deviation, MD, Pattern Deviation, PD).
Both MRF-r and MRF-g returned shorter test times than the HFA (MRF-r 4.7±0.1 (SEM) vs. MRF-g 4.6±0.5 vs. HFA 6.3±0.1 min, P<0.001). Pearson’s correlation R values between MRF-r and HFA for MD found at the four time points ranged from 0.88 to 0.94; and for PD ranged from 0.79 to 0.90. Similarly Pearson’s correlation for MD between MRF-g and HFA ranged from 0.87 to 0.93; and for PD ranged from 0.79 to 0.82. Test-retest repeatability of MRF-r outcomes was high after 2, 4 and 6 months with 95% LoA for MD being 4.0, 5.9, and 5.9 dB respectively. Similar high test-retest repeatability was found using MRF-g, with 95% LoA for MD being 3.7, 5.7, 4.7 dB respectively. This is compared to a more widespread 95% LoA generated by HFA for MD at the same times (9.4, 9.5, 7.4dB respectively). The variance ratios for MRF-r to HFA (F36,36=2.77) and MRF-g to HFA (F36,36 =3.48) confirmed significantly lower (p<0.001) variability with both MRF patterns.
Both radial and grid patterns on MRF gave strong correlations with HFA summary indices across 4 tests over a 6 month time period. Both tests when performed in clinic were shorter and have significantly lower test-retest variability than HFA. MRF has potential to be a portable method of monitoring visual fields.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only