Plasticity and dystonia: a hypothesis shrouded in variability

Abstract

Studying plasticity mechanisms with Professor John Rothwell was a shared highlight of our careers. In this article, we discuss non-invasive brain stimulation techniques which aim to induce and quantify plasticity, the mechanisms and nature of their inherent variability and use such observations to review the idea that excessive and abnormal plasticity is a pathophysiological substrate of dystonia. We have tried to define the tone of our review by a couple of Professor John Rothwell's many inspiring characteristics; his endless curiosity to refine knowledge and disease models by scientific exploration and his wise yet humble readiness to revise scientific doctrines when the evidence is supportive. We conclude that high variability of response to non-invasive brain stimulation plasticity protocols significantly clouds the interpretation of historical findings in dystonia research. There is an opportunity to wipe the slate clean of assumptions and armed with an informative literature in health, re-evaluate whether excessive plasticity has a causal role in the pathophysiology of dystonia.

Keywords: Dystonia; Neurophysiology; Pathophysiology; Plasticity.

Fig. 1

a Plasticity paradigms. Examples of non-invasive brain stimulation (NIBS) techniques that induce plasticity. Response is quantified by taking a mean measure (such as amplitude) of the motor evoked potential (MEP) before and after the session. Protocols that are thought to increase excitability include high frequency repetitive transcranial magnetic stimulation (rTMS), paired associative stimulation with an interstimulus interval of 25 ms (PAS25), intermittent theta burst stimulation (iTBS) and anodal transcranial direct current stimulation (TDCS). Protocols that are thought to reduce excitability are low frequency rTMS, PAS with an interstimulus interval of 10 ms, and continuous TBS (cTBS) and cathodal TDCS. b Variability of paired associative stimulation (PAS25) in writing dystonia is illustrated in 15 individuals. Each dot represents a single patient’s data and the dark line the group mean. Data for both the target muscle, abductor pollicis brevis (APB) and non-target muscle, abductor digiti minimi (ADM) are shown. c Interneuron recruitment and TBS variability. Up to 50% of the variation in TBS was predicted by AP-LM latency, our postulated marker for the efficiency of late I-wave recruitment (see text for detail). Graphs plot the correlation between grand average of cTBS (left) and iTBS effect (right) and AP–LM latency difference