Article Text


Amelioration of spinal myoclonus with levetiracetam
  1. S C Keswani1,
  2. E H Kossoff1,
  3. G L Krauss1,
  4. C Hagerty2
  1. 1Department of Neurology, The Johns Hopkins University, 600 North Wolfe Street, Baltimore, Maryland 21287, USA
  2. 2Neurology Specialists, Columbia, Maryland, USA
  1. Correspondence to:
 Dr G L Krauss;

Statistics from

Spinal myoclonus has been associated with various spinal cord insults, including mass lesions, ischaemia, infection, and as part of a paraneoplastic syndrome.1 It has been postulated that it occurs as a result of deficient inhibitory glycinergic transmission in the spinal cord and subsequent “release” of synchronous motor neurone oscillations within segments of the cord. Levetiracetam (UCB Pharma, Smyrna, Georgia, USA) is a new antiepileptic drug that has been shown recently to reduce the effect of glycinergic inhibitors. We describe three patients whose spinal myoclonus was markedly ameliorated by levetiracetam.


Patient 1: spinal epidural compression

A 62 year old woman with known diffuse large cell lymphoma presented to her oncologist with progressive back pain accompanied by a band-like sensation around her waist. In the preceding four weeks, she had also been troubled by spontaneous involuntary abdominal contractions, and in the preceding two weeks these were accompanied by involuntary jerks of her legs. The patient could not suppress these spontaneous movements; moreover, as voluntary leg movements often precipitated them, she was unable to walk safely because of numerous falls. She denied any limb weakness and bladder or bowel incontinence.

On examination, she had a mild spastic paraparesis with 4+/5 MRC grade power in a pyramidal pattern in the lower extremities (quadriceps, hamstrings, and tibialis anterior), 3+ knee and ankle jerks, and extensor plantar responses bilaterally. There were frequent resting myoclonic jerks of her lower extremities, involving both proximal and distal musculature, occurring at a rate of 150–250/min. There were also occasional, infrequent resting myoclonic jerks affecting the trunk. The myoclonic jerks were exacerbated in amplitude during attempts to perform purposeful movements, suggesting the phenomenonology of action myoclonus. The abnormal movements, rather than weakness, made it impossible for her to stand or walk unassisted. Magnetic resonance imaging (MRI) of the spine revealed malignant infiltration of the lower thoracic vertebrae with evidence of cord compression at T11. An EEG was normal.

She was treated with a maximum tolerated dose of clonazepam (1 g/day) with minimal improvement. She was then started on levetiracetam 250 mg twice daily, and within three days the resting and action myoclonus subsided markedly, such that she was able to walk with no assistance. On examination, the myoclonic jerk frequency in her lower extremities had decreased to 5–10/min, and the jerk amplitude was markedly diminished.

Patient 2: zoster myelitis

An 85 year old woman presented with a three month history of involuntary trunk movements. The movements consisted of sudden extensor jerks of her back. They were spontaneous, occurring several times a day with no obvious provoking factors. Of note, two months before the onset of the movements, she had been diagnosed as having thoracic herpes zoster (at T8) and had subsequent post-herpetic neuralgia. The back movements began as the pain was subsiding. The movements were not painful, but were distressing to the patient as they were socially embarrassing. She was unable to suppress the movements voluntarily. She had been seen by another neurologist who had treated the movements with sodium valproate. She unfortunately received no benefit from this despite a maximum tolerated dose of 2000 mg/day. Past medical history was notable for cardiac arrhythmia and pacemaker placement.

On examination, she had brief, irregular, extensor movements of her thoracic spine, occurring every 10–30 seconds. An EEG was normal. MRI of the thoracic spine was precluded because of her pacemaker. The patient was given levetiracetam at a dose of 250 mg twice a day. Within 24 hours of starting this treatment, the myoclonic movements completely ceased. Two months later, she began to have clusters of repetitive movements once to twice daily for periods of 20–60 minutes. Her dose of levetiracetam was increased to 500 mg twice a day. The movements again ceased, but because of sedation and dizziness at this higher dose, the dosage was reduced to 375 mg/day. At this well tolerated dose, she has been having brief clusters of myoclonic movements two or three times a week.

Patient 3: transverse myelitis

A 12 year old boy presented with a three month history of rhythmic spasms of his right thigh. One month before this symptom, he had had onset of bilateral leg weakness and paraesthesiae and was diagnosed as having acute transverse myelitis. The paraparesis largely resolved within two weeks of onset, but one month later he began having constant, rhythmic jerks of his right quadriceps and hamstrings. These jerks could not be suppressed voluntarily and made walking difficult. Cerebrospinal fluid analysis and an MRI of the spinal cord were normal. An EEG did not show any epileptiform activity. Sodium valproate (1000 mg/d), phenytoin (300 mg/d), and intravenous lorazepam (as often as 2 mg every 4 hours) failed to relieve the constant myoclonus. A trial of botulinum toxin A injections into the right quadriceps did not ameliorate the movements.

On examination, he had constant, semirhythmical myoclonus of his right quadriceps at 120–150 beats/min with his knee extended, and of his right hamstrings with his knee flexed. The myoclonus was not suppressed by patellar fixation, but did improve slightly with concentration on mental tasks. On power testing, there was 4+/5 MRC grade power in the right quadriceps and right hamstrings. A repeat EEG was again unremarkable. He was started on levetiracetam at 250 mg daily and the dose increased over a four week period to 1250 mg/d. No clinical change was noted until the 1250 mg dose was reached, at which point the myoclonus slowed and then completely stopped over a seven day period, allowing independent ambulation. Other than mild initial sedation, no side effects were experienced.


Glycine is a major inhibitory neurotransmitter in the spinal cord, and it has been postulated that deficient inhibitory glycinergic transmission results in dysfunction of segmental spinal cord circuitry, and hence a myoclonic focus in the spinal cord. This postulate is based on studies of animal models of myoclonus2 and an in vitro model of spinal myoclonus.3 The latter study showed that blockade of glycine receptors in isolated spinal cord preparations from neonatal rats enhanced a central pattern generator responsible for 5 to 15 Hz synchronous motor neurone oscillations. Interestingly, these oscillations—generated from as few as two isolated segments—were synchronised over at least six spinal cord segments, suggesting extensive excitatory commissural connections.

It is possible that the effectiveness of levetiracetam in our patients may be related to these glycinergic mechanisms. Levetiracetam has been shown to reverse inhibition of glycine and GABA gated currents induced by negative allosteric modulators, such as zinc and β-carbolines.4 It may therefore conceivably be of benefit in patients with spinal myoclonus by augmenting glycinergic transmission in the spinal cord and thus dampening down myoclonic foci.

In a recent open labelled trial of levetiracetam in eight patients with chronic myoclonus, three of five patients with cortical myoclonus experienced reduction in their myoclonus severity, as assessed by the unified myoclonus rating scale.5 The one patient in this study with spinal myoclonus showed no improvement with levetiracetam. However, the average duration of symptoms in these patients was 7.6 years, ranging from one to 17 years, in contrast to our three patients whose symptoms were one to three months in duration before levetiracetam treatment. It is therefore possible that the differential responsiveness to levetiracetam was because the aforementioned non-responder had a chronic fixed condition whereas our responders had subacute evolving spinal cord injuries.

In a recently published study, levetiracetam was used successfully to treat three patients with posthypoxic and postencephalitic myoclonus, two of whom had failed to respond to valproic acid and clonazepam.6 Add-on therapy with levetiracetam was shown to suppress disabling post-hypoxic cortical reflex myoclonus in a 16 year old boy.7 In another study, severe action myoclonus was suppressed by levetiracetam in three patients, of whom two had Unverricht–Lundborg disease and one had postanoxic myoclonus.8

Our cases, as well as the abovementioned reports of suppression of post-hypoxic and postencephalitic myoclonus with levetiracetam, suggest that this agent is promising for the treatment of both non-cortical and cortical myoclonus. These observations will need to be confirmed in additional patients. Furthermore, the proportion of responders needs to be determined in a larger group of patients, ideally in the setting of a randomised, double blind, placebo controlled trial.


View Abstract


  • Competing interests: GLK is a paid consultant to UCB Pharma. The terms of this arrangement are being managed by the Johns Hopkins University in accordance with its conflict of interest policies. The other authors declare no competing interests.

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.