Long-term PERG monitoring of untreated and treated glaucoma suspects

Purpose: To investigate long-term structural and functional progression of untreated and treated glaucoma suspects (UGS and TGS). Methods: Retrospective analysis of serial steady-state pattern electroretinogram (PERG), mean retinal nerve fiber layer thickness (RNFLT), and standard automated perimetry mean deviation (SAP-MD) in UGS (N = 20) and TGS (N = 18). Outcome measures were the rates of change (linear regression slopes) of PERG amplitude, PERG phase, mean RNFLT, and SAP-MD over 9.8 ± 1.3 years (15.6 ± 4.2 visits). Results: The number of patients with significant (P < 0.05) progression slopes for PERG amplitude, PERG phase, RNFLT, and SAP-MD was, respectively, UGS: 5, 0, 4, 2; TGS: 8, 2, 6, 5. In UGS, outcome measures were not correlated with each other. In TGS, both PERG amplitude and RNFLT were significantly (P < 0.05) correlated with SAP-MD (R ≥ 0.58), while PERG amplitude and RNFLT were not correlated with each other (R = 0.43, P = 0.064). The rate of change of SAP-MD was predicted (P < 0.05) by a linear combination of RNFLT slope and PERG amplitude slope. Conclusions: Results substantiate and extend previous results showing that steady-state PERG amplitude progressively decreased over time in a proportion of glaucoma suspects, with relatively steeper slope in TGS compared to UGS. RNFLT progression also had a steeper slope in TGS compared to UGS; however, progressions of PERG amplitude and RNFLT were not significantly correlated. Both PERG progression and RNFLT progression independently contribute to prediction of visual field progression.