Abstract
Purpose :
Previous work from our lab and others indicates that the pattern electroretinogram (PERG) may be modeled from responses generated by the flash electroretinogram (FERG) using appropriate stimulus patterns. Since patients with primary open-angle glaucoma (POAG) typically exhibit normal FERG responses, but reduced PERG responses, the present work sought to validate this PERG model by testing it in a sample of glaucoma patients.
Methods :
Both transient (2.0 Hz) and steady-state (SS, 7.5 Hz) PERG and FERG responses were collected from a group of patients with diagnoses of severe POAG (N = 15, mean age = 71.60±2.41 years) and age-similar controls (N = 12, mean age = 61.00±3.52 years). The height and width of each stimulus subtended 14.3° ×14.3°. To create the PERG simulation, long-duration increment and decrement FERG responses were additively combined and then subjected to a series of modeling parameters that manipulated the amplitudes and phases of the individual responses to the increment and decrement flashes. Amplitudes were measured from both the actual PERG responses and the simulations for both the control group and the patient group.
Results :
For PERG recordings, amplitude of the transient P50 component was found to be statistically equal between the control group and the POAG group. However, N95 amplitude was reduced in the POAG group relative to the control group (M = 1.198 vs. 2.201; t(25) = 2.522, p = 0.018), and this same pattern was found for SS amplitudes of the POAG group relative to the control group (M = 0.573 vs. 0.347, t(25) = 2.621, p = 0.015). Using the modeling parameters that provided the best fit for each individual, the same pattern was found with the P50 showing no significant amplitude difference between the POAG and control group, but both N95 and SS amplitudes being reduced in the group with POAG relative to the controls (N95: M = 1.385 vs. 2.205, t(25) = 2.722, p = 0.012; SS: M = 0.322 vs. 0.438, t(25) = 2.372, p = 0.026). This suggests that both N95 and SS PERG responses can be successfully modeled using long-duration FERG responses.
Conclusions :
Both the N95 and steady-state amplitudes from simulations could be adequately modeled in POAG patients and age-similar controls. Further studies with larger sample sizes will be required to address the predictive validity of PERG modeling as a tool for tracking disease progression in clinical populations.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.