Article Text

Motor cortex stimulation does not reset primary orthostatic tremor
  1. Clinical Neurophysiology Unit, University Department of Clinical Neurology, The Radcliffe Infirmary, Oxford OX2 6HE, UK
  1. Dr KR Mills.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Primary orthostatic tremor is a rare but well characterised condition.1-3 The predominant feature is unsteadiness and shaking of the legs, which occurs exclusively on standing. Electromyographic recordings disclose a 13-18 Hz tremor in leg and paraspinal muscles.2 3 A similar tremor in the arms can be induced by weight bearing actions.

We studied five patients with previously established clinical and EMG diagnoses of primary orthostatic tremor. Surface EMG recordings were made from up to four muscles simultaneously; one muscle (right hamstrings) was recorded from throughout each study to act as a timing reference. Leg muscles were studied with the patients standing and arm muscles with patients leaning forwards on their arms. The effect of different maintained head positions (looking to the left, to the right, upwards, and downwards) was also recorded.

In three patients, while standing, transcranial magnetic stimulation (TMS) was applied using a Magstim 200 with a double cone coil positioned over the vertex. Stimulus intensity was sufficient to produce consistent motor evoked potentials in voluntarily activated leg muscles while seated. The magnetic stimuli were triggered, at a maximum of 0.3 Hz, by EMG bursts. The stimulus delay after an EMG burst was varied between 0 and 50 ms using a precise timing device, to see whether any effect of TMS was dependent on its temporal relation to the tremor cycle. Ten stimuli were applied at each delay setting.

The EMG signals were digitised and data were rectified and displayed in sweep durations of 200-500 ms. The 10 individual trials for a given delay were averaged, using the stimulus as a reference point. In addition all interburst intervals in each individual trial were measured. This allowed calculation of any latency shift in the first poststimulus tremor burst, as well as any subsequent alteration in tremor frequency.

The frequency of tremor ranged between 14 and 18 Hz. Leg tremor was present in all patients on standing and disappeared on sitting, whereas arm tremor was only seen when patients leant on their arms. Within each patient there was a clear and consistent phase relation of tremor bursts between different muscles. However, comparing all five patients the exact timing of these relations varied. Specifically, there was no overall relation between flexor and extensor muscles or between left and right sides.

We found that changes in head position while standing had no effect on either the frequency or amplitude of the tremor. This would suggest that static vestibular inputs to spinal motor neurons do not modulate the tremor.

In two patients TMS reduced the amplitude of the next EMG burst after the evoked compound muscle action potential (CMAP), but did not alter the frequency of the tremor (figure, A). The magnitude of reduction in amplitude was variable and did not correlate with alterations in stimulus delay. Motor neurons discharged as a result of TMS of the motor cortex would be expected to remain partially refractory and this would explain the reduction in burst amplitude.

All traces show rectified and averaged (10 trials) EMG data over a sweep duration of 500 ms. (A) and (B) refer to different patients, recording from right quadriceps and right hamstrings respectively. In both cases the upper trace is from a recording without transcranial magnetic stimulation. (A) The stimulus was triggered 45 ms after the start of an EMG burst. The EMG burst after the CMAP is reduced in amplitude, but occurs at the expected latency (vertical line). (B) The stimulus was triggered at the start of an EMG burst. EMG bursts disappear after the CMAP, but then reappear at the expected latency (vertical line).

In the third patient, EMG bursts were transiently suppressed for between one and three tremor cycles after TMS (figure, B), before reappearing at the expected latency and without any subsequent modulation of frequency. There was no correlation between degree of suppression and stimulus delay. This suppression can probably be attributed to the “silent period”, which is thought to be due to activation of both intracortical and spinal inhibitory mechanisms. Significantly, these inhibitory mechanisms did not seem to have modulated the tremor generator, as the EMG burst reappeared at the expected time and the tremor continued at the prestimulus frequency.

The generator site underlying primary orthostatic tremor is unknown. The frequency of the tremor (13-18 Hz) suggests that it is central rather than peripheral. The failure of peripheral stimuli to reset primary orthostatic tremor2 supports this. The effect of transcranial magnetic stimulation, which has been reported to reset both essential tremor and parkinsonian tremor,4 has not previously been reported on in this condition.

We suggest that the motor cortex is neither the site nor a modulator of the presumed central generator of primary orthostatic tremor. Furthermore, other intracortical structures activated by transcranial magnetic stimulation (for example, intracortical or transcallosal inhibitory pathways) do not seem to modulate the generator.


This research was funded in part by a Wellcome Trust research grant (KAN).