Reproducibility of pattern electroretinogram in glaucoma patients with a range of severity of disease with the new glaucoma paradigm

Purpose: To determine the reproducibility of the pattern electroretinogram with the new Pattern Electroretinogram for Glaucoma (PERGLA) recording paradigm in glaucoma patients with a range of severity. Design: Experimental study. Participants: Fifty-three glaucoma patients were recruited for the study (mean age +/- standard deviation [SD], 69 +/- 11 years). Their mean deviation (MD) global indices on static automatic perimetry ranged from 2.16 to -31.36 decibels (mean MD, -9.05). Intervention: All patients had pattern electroretinogram recordings done 5 times by the same operator, on 5 different days with the standardized PERGLA paradigm. Main Outcome Measures: Pattern electroretinogram amplitude (microvolts), phase (pi radians), response variability (coefficient of variation [CV] = SD/mean x 100) of amplitude and phase of 2 partial averages that build up the pattern electroretinogram waveform, interocular asymmetry in amplitude and phase (in terms of the CV generated by the pattern electroretinogram software), signal-to-noise (S/N) ratio, SDs, CV, and intraclass correlation coefficient (ICC). All analyses were done on one eye of each subject, except when interocular asymmetry was studied. Results: The CVs of intrasession variabilities in amplitude and phase were 12.08% and 2.20%, respectively, and those of intersession variabilities were 20.82% and 4.17%. The pattern electroretinogram produced intersession ICCs in amplitude and phase of 0.791 and 0.765, respectively. These ICCs were significantly higher than the ICCs for pattern electroretinogram interocular asymmetry in amplitude and phase (0.659 [P<0.05] and 0.571 [P<0.05], respectively). On average, the pattern electroretinogram S/N ratio in glaucomatous patients was about 5:1. Conclusions: The reproducibility of PERGLA in glaucomatous patients is sufficiently good for it to be considered a useful complementary clinical tool. Being more reproducible, direct measures of amplitude and phase should be more useful in monitoring progression than interocular asymmetry comparisons.