Evaluation of Retinal Nerve Fiber Layer Thickness, Electroretinogram and Visual Evoked Potential in Patients with Alzheimer’s Disease

Abstract

Objective. To study the retinal nerve fibre layer (RNFL) thickness and visual electrophysiology testing in patients with Alzheimer’s disease (AD). Methods. A cross-sectional, hospital-based study: 25 AD subjects and 25 controls were recruited. Candidates who fulfil the criteria with normal ocular examinations were made to proceed with scanning laser polarimetry, pattern electroretinogram (PERG), and pattern visual evoked potential (PVEP) examinations of the right eye. RNFL thickness, PERG, and PVEP readings were evaluated. Results. In AD, the mean of average RNFL thickness was 45.28 μm, SD = 3.61, P<0.001 (P<0.05), while the superior RNFL thickness was 54.44 μm, SD = 2.85, P<0.001 (P<0.05) and inferior RNFL thickness was 47.11 μm, SD = 4.52, P<0.001 (P<0.05). For PERG, the mean P50 latency was 63.88 ms, SD = 7.94, P<0.001 (P<0.05) and the mean amplitudes of P50 waves were 1.79 μV, SD = 0.64, P<0.001 (P<0.05) and N95 waves were 2.43 μV, SD = 0.90, P<0.001 (P<0.05). For PVEP, the mean latency of P100 was 119.00 ms, SD = 9.07, P<0.001 (P<0.05), while the mean latency of N135 was 145.20 ms, SD = 8.53, P<0.001 (P<0.05). The mean amplitude of P100 waves was 3.71 μV, SD = 1.60, P<0.001 (P<0.05), whereas the mean amplitude of N135 waves was 3.67 μV, SD = 2.02, P<0.001 (P<0.05). RNFL thickness strongly correlates with PERG readings, with P50 latency R = 0.582, R2 = 0.339, P=0.002 (P<0.05), amplitude of P50 wave at R = 0.749, R2 = 0.561, P≤0.001 (P<0.05), and amplitude of N95 wave at R = 0.500, R2 = 0.250, P=0.011 (P<0.05). No significant difference and correlation were observed on PVEP readings. Conclusion. The mean of the average, superior and inferior RNFL thickness were significantly lower in the AD group compared with control. There is also significant difference of PERG and PVEP parameters between AD and controls. Regression analysis showed average RNFL thickness having significantly linear relationship with the PERG parameters.