Semi-invasive and non-invasive recording of visual evoked potentials in mice

Purpose: Visual evoked potentials (VEPs) are used to assess visual function in preclinical models of neurodegenerative diseases. VEP recording with epidural screw electrodes is a common method to study visual function in rodents, despite being an invasive procedure that can damage the tissue under the skull. The present study was performed to test a semi-invasive (epicranial) and a non-invasive (epidermal) VEP recording technique, comparing them with the classic epidural acquisition method. Methods: Flash VEPs were recorded from C57BL/6 mice on three separate days within 2 weeks. Waveforms, latencies and amplitudes of the components were compared between the three different methods, utilizing coefficient of repeatability, coefficient of variation and intersession standard deviation to evaluate reproducibility. Results: While epidural electrodes succeeded in recording two negative peaks (N1 and N2), epicranial and epidermal electrodes recorded a single peak (N1). Statistical indexes showed a comparable reproducibility between the three techniques, with a greater stability of N1 latency recorded through epicranial electrodes. Moreover, N1 amplitudes recorded with the new less-invasive methods were more reproducible compared to the invasive gold-standard technique. Conclusions: These results demonstrate the reliability of semi- and non-invasive VEP recordings, which can be useful to evaluate murine models of neurological diseases.