Flash visual evoked potentials are unreliable as markers of ICP due to high variability in normal subjects

Abstract

Previous publications have suggested a high correlation between flash visual evoked potential (F-VEP) N2 peak latency and intracranial pressure. This would enable F-VEP to be used as a non-invasive and inexpensive method to estimate ICP in a number of settings. However, basic knowledge about variability across subjects and test-retest properties of the F-VEP is lacking. Fifteen healthy adult subjects were tested on three different occasions. F-VEP responses were recorded using international standards. For the tested population, mean N2 latency was 65.7 ms (SD 10.7 ms) and the range was 48-110 ms. Intra-individual variability was high, in four of the 15 subjects more than 15 ms between testing sessions. The same was found for P2 latency and for N2 and P2 amplitudes. The response waveform was very variable and unambiguous marking of peaks was often difficult. One out of the 15 subjects had a very poorly developed F-VEP response, but a normal pattern-reversal VEP response. F-VEP has a wide range of latency, amplitude and waveform across normal subjects. A large proportion of subjects also had a high intra-individual variability over time. This variability makes F-VEPs unreliable as a marker for intracranial pressure, and caution in interpreting F-VEP changes in clinical work is advised.