Diagnostic performance of multifocal photopic negative response, pattern electroretinogram and optical coherence tomography in glaucoma

The photopic negative response of the electroretinogram reflects retinal ganglion cell function and consequently aids diagnosis of optic nerve diseases including glaucoma. In this study, we assessed the efficacy of stimulation parameters for electroretinographic recordings of the multifocal photopic negative response (mfPhNR) for the detection of glaucoma and compared the diagnostic accuracy of electrophysiological, structural and functional measures of glaucoma. We compared the diagnostic performance of the mfPhNR for 6 different stimulation rates in a cohort of 24 controls, 10 glaucoma suspects (GLA(S)) and 16 glaucoma participants (GLA(G)). A cross-modal comparison of the mfPhNR/b wave ratio was performed with the pattern electroretinogram (PERG), and the peripapillary retinal nerve fiber layer (pRNFL) thickness. These analyses were based on area under curves (AUC) obtained from receiver-operating-characteristics (ROC) and step-wise regression analyses. We found that compared to the other mfPhNR-conditions, the PhNR/b-wave ratio for the fastest stimulation condition had the highest AUC for GLA(S) (0.84, P = 0.008, 95%CI: 0.71- 0.98), while the other modalities, i.e., PERG-amplitude and pRNFL had AUCs of 0.78 (P= 0.039), and 0.74 (P < 0.05), respectively. For GLA(G), the respective AUCs were 0.78 (P= 0.004), 0.85 (P< 0.001) and 0.87 (P< 0.001). pRNFL was the significant predictor for both mfPhNR/b-wave ratio [t (48) = 4, P = 0.0002] and for PERG amplitude [t (48) = 3.4, P = 0.001]. In conclusion, fast mfPhNR protocols outperform other multifocal PhNR protocols in the identification of glaucomatous damage especially for GLA(S) and thus aid the early detection of glaucoma, indicating its value as a surrogate marker of early stage ganglion cell dysfunction.