Purchase this article with an account.
Jonathan C. Clarke, Trishal Boodhna, Ananth Viswanathan; The Progressor Database: Studying Trends In Initial Visual Field Progression And Reliability. Invest. Ophthalmol. Vis. Sci. 2011;52(14):4157. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate whether initial visual field (VF) progression and reliability indices have changed over a defined time period in subjects attending a tertiary glaucoma service
138,597 visual fields dated from 1989 to 2010 were sourced from the PROGRESSOR database of the Glaucoma Service of Moorfields Eye Hospital, London. VF progression rates were compared in order to assess changes in initial VF deterioration in recent years (pre-2005, n=5322 compared post-2005, n=3753) at Moorfields Eye Hospital and the impact of changes in test strategy from full threshold to SITA on perimetric reliability indices was examined.
The decline in mean sensitivity (MS) and mean deviation (MD) from the first to the second VF was less severe post 2005 than pre 2005. The decline in MS was 0.16 dB (P=0.003, ANOVA) less severe and the decline in MD was 0.14 dB (P=0.05, ANOVA) less severe and these findings were repeated between the second and third and third and fourth VF’s. Mean inter-test interval was 370 (±5.86) days pre-2005 and 282 (±13.45) days post-2005. Significant differences in reliability indices were observed between full threshold and SITA. False negatives were found to be significantly lower with SITA (-5%, P<0.000) whilst fixation losses were significantly lower with the full threshold strategy (7%, P<0.000)
Small but significant ameliorations were found in initial VF decline post-2005 compared to the pre-2005 era and these may be partially explained by the reduction in interest intervals post-2005 compared to pre-2005. Significant differences in reliability criteria were found between Full Threshold and SITA and these may possibly be related to the differences in ascertainment and calculation between the two algorithms.
This PDF is available to Subscribers Only