Reproducibility in the global indices for multifocal visual evoked potentials and Humphrey visual fields in controls and glaucomatous eyes within a 2-year period.

PURPOSE: In previous studies, we applied receiver operating characteristic curve analysis to the signal-to-noise ratio distributions in the signal and noise windows of multifocal VEP (mfVEP) response. The areas under the curve thus obtained (SNR-AUC) were found to quantitatively detect glaucomatous visual field damage. The present study evaluated the reproducibility of SNR-AUC and the Humphrey visual field (HVF) global indices in 37 eyes with primary open angle glaucoma (POAG; POAG group) and in 30 controls (control group) within a 2-year period. METHODS: The HVF SITA standard 24-2 and mfVEP were recorded at three separate sessions for each individual. The intersession variability for SNR-AUC, mean deviation (MD), and pattern standard deviation (PSD) was evaluated using the repeated measures of analysis of variance and Bland-Altman plots. The logarithmically converted coefficients of variation (CV) of PSD and SNR-AUC were compared between the control and POAG groups. Linear regression analyses were performed on the logarithmic CV of SNR-AUC against the average MD, PSD, and SNR-AUC. RESULTS: SNR-AUC in the POAG group was significantly lower and its CV was greater compared with the control group (P < 0.0001). MD value recorded at the third visit had significantly improved than that at the first visit in the control group (analysis of variance, P = 0.03), whereas PSD value was significantly worse in the POAG group (P = 0.024). In the POAG group, SNR-AUC CV increased as the glaucoma stage became more advanced when evaluated by any functional parameters tested (i.e., MD, PSD, or SNR-AUC). CONCLUSIONS: The SNR-AUC of mfVEP showed a high reproducibility in control group, whereas it fluctuated more in the POAG group according to the disease severity. MD in the control group and PSD in POAG group fluctuated among sessions during the 2-year period.

Effect of refractive errors on multifocal VEP responses and standard automated perimetry tests in a single population.

BACKGROUND: We have previously reported that the degree of signal-to-noise ratio (SNR) distribution overlaps between a signal window and a noise window in multifocal VEP (mfVEP) responses, which is determined by the area under the receiver-operating characteristic curve termed SNR-AUC, can quantitatively detect glaucomatous visual functional damage. However, the effect of high myopia on this parameter is not yet known. METHODS: SNR-AUC, total deviation, and retinal sensitivity on the Humphrey visual field (HVF) test were compared among 34 eyes>-6 diopters (control) and 21 eyes≤-6 diopters (high myopia), both of which were ophthalmoscopically normal and had a best-corrected visual acuity of 20/20. The mfVEP and HVF parameters were obtained from stimulus areas that corresponded to both HVF 24-2 and 10-2 programs. RESULTS: Both the HVF 24-2 total deviation and the SNR-AUC obtained from 60 sectors in high-myopia patients were significantly lower compared with controls (P=0.045 and P=0.003, respectively). The SNR-AUC obtained from the central 36 sectors that corresponded to the HVF 10-2 area in high-myopia patients was also significantly lower than that of the controls (P=0.01). Multiple regression analyses demonstrated that age and refractive error were significantly associated with retinal sensitivity on the HVF 24-2 and SNR-AUC for the whole field and central field, respectively. CONCLUSIONS: High myopia reduces the SNR-AUC of mfVEP responses, even with refractive correction. A normative database should be separately established for high myopes to evaluate the mfVEP responses obtained from highly myopic glaucoma patients.