Paired-pulse behavior of visually evoked potentials recorded in human visual cortex using patterned paired-pulse stimulation.

Paired-pulse stimulation techniques are used as common tools to investigate cortical excitability and cortical plastic changes. Similar to investigations in the somatosensory and motor system here we applied a new paired-pulse paradigm to study the paired-pulse behavior of visually evoked potentials (VEPs) in 25 healthy subjects. VEPs were recorded and the responses to the first and the second P100 peak were analyzed at different SOAs [stimulus onset asynchrony (SOA) = interstimulus interval (ISI) + pulse duration (13 ms)]. Two measures describe the paired pulse interaction: the "amplitude ratio", the ratio of the second to the first amplitude, and the "latency shift", the difference of the inter-peak interval between the P100 peaks and the respective SOA. To separate alterations in the amplitude of the second VEP response due to changes in paired-pulse inhibition from those originating from superposition of the two waveforms, particularly at short SOAs, we created a waveform template from recordings made at SOAs of 1 s, where interaction can be assumed to be negligible. Superposed traces of VEP recordings were then created by adding two templates at delays corresponding to the SOAs used. The original recordings were then digitally subtracted from the traces obtained by superposition. Analysis of the subtracted traces revealed evidence that at short SOAs the second VEP response is substantially suppressed, a finding comparable to the paired-pulse inhibition described for motor and somatosensory cortex following paired-pulse stimulation. However, paired-pulse inhibition seen in V1 varied considerably from subject to subject, both in respect to amplitude, and to time of maximal inhibition. We found paired-pulse inhibition ranging from 12 to 76% (mean 34%) at SOAs between 80 (shortest discriminable SOA) and 320 ms (mean 128 ms). At intermediate SOAs between 80 and 720 ms (mean 215 ms) the amplitude ratios were between 94 and 145% (mean 116%) indicative of slight paired-pulse facilitation. Comparable to recovery studies by means of paired-pulse median nerve stimulation in somatosensory cortex, at shorter SOAs we found a delayed second VEP response. Combined together, our findings suggest that VEPs are characterized by significant paired-pulse inhibition at short SOAs, a phenomenon reminiscent of findings reported in other modalities. Possible mechanisms and pharmacological properties of the described paired-pulse behavior in visual cortex remain to be explored.