TMS for Dystonia
Transcranial magnetic stimulation (TMS) has been tried in dystonia for about twenty years. This page summarizes what was aimed where, in whom, and what happened, with a focus on cervical dystonia.
How this was made. It is a research summary, not medical advice. It covers 18 studies found through PubMed searches and read at the abstract level (their titles link to PubMed), so it is a sample of the literature and not a complete review. Numbers are as the abstracts report them. The results are mostly from small studies.
For the wider research picture see the
Dystonia Research Tracker and the
CD Crash Course.
The bottom line
- TMS is still experimental for dystonia. The treatment studies here had roughly 6 to 16 people each, and the search turned up no large randomized trial showing a lasting, clinically meaningful benefit. A 2023 review reaches the same view: small samples, mixed populations, different targets and inconsistent control arms make a firm conclusion hard.
- Best signal so far in cervical dystonia: the cerebellum. Two weeks of cerebellar theta-burst stimulation gave about a 15% improvement on the TWSTRS severity scale versus sham (Koch 2014, Rome), and a 10-day version paired with neck training gave small severity gains and larger, longer-lasting quality-of-life gains (Bradnam 2016, Sydney).
- Newest signal: the somatosensory cortex. In writer's cramp, 10 Hz stimulation of the primary somatosensory cortex improved writing dysfluency in a 12-person double-blind crossover, and premotor stimulation did not (Bukhari-Parlakturk 2025, Duke). In cervical dystonia, somatosensory stimulation has so far been shown to engage the brain and change physiology, not to improve symptoms.
- Many results are not about symptoms. Seven of the 18 studies below measured brain physiology or connectivity, not symptoms. They are useful for choosing targets but don’t show a treatment works.
- Standard care hasn’t changed. Botulinum toxin remains the mainstay for cervical dystonia, with deep brain stimulation for severe cases; TMS would be a research option at most.
What TMS is, in a paragraph
A coil held against the scalp delivers brief magnetic pulses that painlessly induce a small electric current in the brain region beneath it. Repeated pulses (rTMS) can shift how excitable that region is for a while afterwards. As general background: slow rates (about 1 Hz, or 0.2 Hz) are usually used to dampen a region; faster rates (10 Hz) to raise it; theta-burst protocols deliver bursts at a rapid pattern (continuous, cTBS, is usually treated as dampening; intermittent, iTBS, as boosting). The dystonia rationale is that people with dystonia often show reduced intracortical inhibition, a weakened “brake” in the motor cortex, and that the condition involves a network including the cerebellum, basal ganglia and sensory cortex, so several different targets have been tried.
TMS is also used as a measuring tool: single pulses can record inhibition (for example the cortical silent period) without treating anything, and many of the studies below did exactly that.
Where the coil has been aimed
The studies
Condition
Outcome
Reading the numbers. TWSTRS is the standard cervical dystonia severity scale; a lower score is better, so a negative change means improvement. “Sham” is a fake stimulation used as the comparison. A result in 8 to 16 people, in one session or one week, can easily be chance or placebo response, which is why the authors themselves usually call these findings preliminary.
Who is doing this work
Institutions are as PubMed records them for the first author of each study above, in the year of publication. Some of the groups are the same researchers as on your MRI pages: the Minnesota laryngeal-dystonia TMS work is Prudente’s, and the Kokkonen stimulation target came from the lesion-network paper that Prudente co-authored.
What is still unknown
- Best target. Cerebellum, premotor cortex, motor cortex and somatosensory cortex have each had a small positive signal, and no head-to-head trial compares them in cervical dystonia.
- Protocol. Frequency, number of pulses, sessions per week and total duration all vary. The somatosensory result that worked used 10 Hz, not the dampening low-frequency approach used in most earlier work.
- Durability. Most studies measured immediately after stimulation; only a few followed people for weeks (Bradnam kept quality-of-life measures to 12 weeks).
- Who responds. Kimberley’s hand-dystonia study found age predicted response; the 2024 Turku study linked the brain response to severity and to how well a person’s sensory trick works.
- Combining with usual care. The 2023 review suggests TMS may be better used alongside standard treatment, which no trial here tested in cervical dystonia.