Do alternate methods of analysing motor evoked potentials give comparable results?

This study assessed the reliability of alternate methods of analysis of motor evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS). We recorded two sets of MEPs (Time 1 and Time 2) at the optimal scalp sites for both the right first dorsal interosseous (FDI) and flexor carpi ulnaris (FCU) at two different stimulation intensities in 10 healthy subjects. MEP magnitude was determined in each of the following three ways: the mean peak-to-peak amplitude and area of the 20 individual responses; the amplitude and area of the ensemble averaged waveform; and the amplitude and area of the maximal response. There was no significant difference in amplitude or area for either muscle using any of the three methods between Time 1 and 2. However, the ensemble average (area and amplitude) was significantly smaller that the mean MEP, and the maximal MEP amplitude was significantly larger. Intraclass correlation analysis demonstrated that reliability of MEP measures over time was poor regardless of method. Reliability was similar between methods for FDI, but FCU had lower reliability values for the mean and ensemble average methods than the maximal method.