To compare central visual field progression using mean deviation and pointwise linear regression analysis

PURPOSE: To compare central visual field progression using mean deviation and pointwise linear regression (PLR) analysis. METHODS: We analyzed the 10-2 Humphrey visual field (HVF) tests for moderate and advanced primary glaucoma who had undergone at least five reliable 10-2 visual field tests with a minimum follow-up of at least two years and best-corrected visual acuity better than 6/12. Regression slope less than −1 dB/year at P < 0.01 at a point was defined as an individual threshold point progression. RESULTS: Ninety-six eyes of 74 patients were included. The median follow-up duration was of 4 years (±1.97). Median 10-2 mean deviation (MD) at inclusion was −19.01 dB (interquartile range [IQR] −13.2, −24.14) and −21.90 (IQR − 13.4, −27.8) on 24-2 HVF. The median rate of MD change was −0.13 dB/year (IQR − 0.46, 0.08) for 10-2. The median rate for visual field index (VFI) change was 0.9% per year (IQR − 1.5, 0.4). Twenty-eight percent of eyes (27 eyes) showed progression. Twelve percent (12 eyes) showed progression of two or more points in the same hemifield on pointwise linear regression (PLR) analysis, and 16% of eyes (15 eyes) showed progression of one point. The median rate of MD change was significantly more in progressing eyes based on PLR analysis than eyes with no progression (−0.5 vs. −0.06 dB/year P < 0.001). One patient had likely and the second had possible progression on 24-2. In 24 eyes, event analysis did not show any change; the rest mean deviation was out of range. CONCLUSION: Central visual field PLR analysis is useful in detecting progression in advanced glaucomatous damage.