Plasticity and intracortical inhibition in dystonia--methodological reconsiderations.

Sir, We would like to thank Dr Rosenkrantz for her interest in our recent paper on physiological differences between patients with psychogenic and organic forms of limb dystonia. We are also very grateful for the opportunity to discuss in detail the connection between input–output relationships and effects of paired associative stimulation (PAS). It is a topic of some confusion among those who use transcranial magnetic stimulation and it is worthwhile exploring in detail possible connections between these two phenomena. In the usual PAS experimental design, a standard transcranial magnetic stimulus is used to evoke motor evoked potentials in the abductor pollicis brevis muscle. Application of the PAS25 protocol increases the amplitude of motor evoked potentials evoked by this standard pulse for the following 30 min or so. In contrast, the input–output curve uses a range of transcranial magnetic stimulation intensities to plot the relationship between intensity and size of evoked motor evoked potential. As Dr Rosenkranz points out, if the effect of PAS25 is equivalent to increasing the intensity of the transcranial magnetic stimulation pulse, then subjects who have a steeper input–output curve will achieve a greater increase in amplitude of motor evoked potential than those with a shallow input–output curve. However, first it is important to underline that proposing PAS25 to be the same as increasing the intensity of the transcranial magnetic stimulation pulse is not equivalent as saying that the effect is the ‘same’ as increasing transcranial magnetic stimulation intensity. Measurements of the strength-duration properties of transcranial magnetic stimulation indicate that it stimulates axons; turning up stimulation intensity increases response size because more axons are recruited at higher intensities. For PAS25 to have the ‘same’ effect we would have to propose that PAS25 increases the excitability of axonal membranes, so that more axons are stimulated by the standard transcranial magnetic stimulation pulse after a PAS25 protocol. This may be a remote possibility, but seems unlikely in view of the fact that there is good evidence that PAS25 involves changes at glutamatergic synapses. Indeed a much more consistent hypothesis is that PAS25 does not affect the number of axons stimulated, but instead it increases the effectiveness of the terminal synapses attached to those axons. This then could result in a greater total depolarization of a second set of neurons and a larger motor evoked potential response. In this case, from the point of view of the receiving neurons, the effect of PAS25 could appear equivalent to that of increasing the transcranial magnetic stimulation intensity: in each case they receive more depolarization. So how does this increased depolarization interact with different input–output slopes? The answer depends on the reason why the input–output slopes differ between groups of individuals. Let us imagine two groups of subjects, one of which has a steeper input–output slope than the other. The slope of the input–output curve depends on the distribution of excitability at all levels of the corticospinal pathway. It could, for example, be increased because of an increase in spinal excitability. That is, the amount of depolarization reaching the spinal cord after a given stimulus could be the same in two groups of subjects, but in the group with higher spinal excitability this would discharge more spinal motoneurons than expected and the input–output slope would be steep. In this case, the effect of PAS25 on the cortex doi:10.1093/brain/awq025 Brain 2010: 133; 1–3 | e147