Purpose:
1. To compare the proportion of progressing eyes and the time to progression, determined by the GDx progression analysis program (GPA), NFI change (reproducibly exceeding 35 or increasing ≥ 7), and conversion on VF (VF change), in initially perimetrically unaffected eyes of glaucoma patients with monocular visual field loss. 2. To explore the agreement in eye classification (progressing vs. non-progressing) between methods.
Methods:
In 34 primary open-angle glaucoma patients, we performed 6 monthly GDx VCC imaging and standard automated perimetry (SAP) over a period of 51.6 (SD = 0.7) months.
Results:
Eleven eyes (32.4%) were classified by 1 of 3 methods as progressing, whereas 5 eyes (45.5%) were classified by all 3. With GPA, 20.6% (7 eyes) showed progression; with the NFI change and VF change, this was 23.5% (7 eyes) and 20.6% (8 eyes), respectively. These percentages were similar for the 3 methods (p = 1.0, pair-wise comparisons). No differences in time to progression between these 3 methods were observed (p = 0.829). At baseline, 44.1% (15/34) of the eyes already had an NFI ≥ 35, of which 20% (3/15) showed progression during the follow-up (2 with all 3 methods and 1 with GPA and NFI change only). Of the eyes with an NFI < 35 at baseline, 26.3% (5/19) progressed.
Conclusions:
Although the proportion of progressing eyes in the initially perimetrically unaffected eyes of glaucoma patients with monocular VF loss did not differ between the 3 methods, not all methods identified the same eyes as progressing. Structural change detection was at least as sensitive as functional change detection; in many eyes it was even more sensitive.
Keywords: retina • optic nerve • imaging/image analysis: clinical