Myoclonus in Epilepsy Patients with Anticonvulsive Add-On Therapy with Pregabalin


Address correspondence and reprint requests to Dr. H.-J. Huppertz at Epilepsy Center, University of Freiburg, Breisacher Str. 64, D-79106 Freiburg, Germany. E-mail:


Summary:  Purpose: To report on the occurrence of myoclonus in patients receiving pregabalin (PGB) for the treatment of focal epilepsy.

Methods: Clinic records of 19 patients who were consecutively enrolled at a tertial referral epilepsy center in a randomized, double-blind and/or open add-on study with PGB were reviewed.

Results: In four patients treated with PGB, focal myoclonus newly developed. The side effect appeared with PGB doses of 50–600 mg/day; the intensity showed some dose dependency. All patients had medically refractory focal epilepsy and received other antiepileptic drugs (AEDs) besides the study medication. One patient showed focal myoclonic jerks of the left arm, whereas the other patients developed multifocal myoclonus. Polygraphic studies including electromyogram (EMG)-triggered back-averaging of the EEG in the patient with the highest frequency of myoclonic jerks showed no visible correlate of the myoclonus. In this patient, frequency and intensity of myoclonic jerks significantly decreased after dose reduction of PGB. In the other cases, myoclonus was only subtle and did not significantly interfere with daily activities, so that a dose reduction of PGB was not considered necessary.

Conclusions: These data indicate a relatively high incidence (four of 19) of myoclonus associated with PGB therapy. The rate seems to be at least as high as reported in patients receiving the structurally similar anticonvulsant gabapentin.

Myoclonus appears in a wide variety of diseases such as epilepsy, posthypoxic brain damage, metabolic derangement, and focal brain lesions. In addition, it has been observed as side effect of medication, for example of antiepileptic drugs (AEDs) like gabapentin (GBP) (1,2), lamotrigine (LTG) (3,4), carbamazepine (CBZ) (5–7), and vigabatrin (VGB) (8). Our report refers to the occurrence of myoclonus in patients receiving pregabalin (PGB), a newly developed AED against partial seizures, which is structurally similar to GBP and whose efficacy, dose–response characteristics, and potential adverse effects are presently evaluated in clinical studies. Like GBP, PGB [isobutyl γ-aminobutyric acid (GABA)] (CI-1008) is an amino acid structurally related to the neurotransmitter GABA, but inactive at GABA receptors and transporters. The mode of action is not well understood; proposed mechanisms are an increased rate of GABA synthesis, an alteration of nonsynaptic GABA release, and a reduction of neuronal calcium currents by an interaction with the auxiliary α2δ subunit of voltage-sensitive calcium channels (9). Recently evidence has been presented for an agonism of gabapentin at adenosine triphosphate (ATP)-dependent potassium channels (10). Our own results (to be published) suggest also that PGB acts as agonist at KATP channels of the plasma membrane. As far as is known, the profile of potential adverse effects of PGB is similar to that of GBP, whereas the antiepileptic efficacy seems to be stronger (11). The drug is planned to be introduced in the United States in 2001.


Clinic records of 19 patients with medically refractory focal epilepsy receiving an add-on therapy with PGB at a tertiary referral epilepsy center (Freiburg, Germany) were reviewed. Whereas three patients participated in a randomized, double-blind study (150 or 600 mg PGB/day, or placebo) before entering the open add-on phase, the others were directly enrolled in the latter study (PBG at a dose range of 25–600 mg/day). In one patient who developed a skin rash after 2 days of treatment, the medication was immediately stopped during the double-blind study and not restarted again. All patients were treated by the same investigator (H.-J.H.) and were specifically asked for the occurrence of any side effects including myoclonus before and during treatment with PGB. Myoclonus was defined as sudden, brief, monophasic, shock-like, involuntary contractions of a group of muscles, which are irregular in rhythm and amplitude. When appearing asynchronously in two or more limbs, it was classified as multifocal; otherwise it was focal.


Four of 19 patients newly treated with PGB developed myoclonus. The clinical features are shown in Table 1. Two patients had symptomatic partial epilepsy due to left hippocampal atrophy and viral meningoencephalitis in childhood, respectively. In the other patients, no etiology could be established from MR images or history. No patient had a history of myoclonic jerks before PGB treatment. However, patients 1 through 3 had tonic and clonic elements limited to some part or one side of the body during their partial seizures.

Table 1.  Clinical features of patients with pregabalin-associated myoclonus
Etiology of epilepsySeizure typeMyoclonus
PGB dose per dayOther AED
Time course
  1. SPS, simple partial seizures; CPS, complex partial seizures; GTCS, secondarily generalized tonic–clonic seizures; CBZ, carbamazepine; PB, phenobarbital; CLB, clobazam; LTG, lamotrigine; PGB, pregabalin.

1M44Viral meningoencephalitisSPS, CPSMultifocal (both arms + legs), high frequency600 mg (8.96 mg/kg)CBZ 2,000 PB 125Myoclonus ↓ with reduction of PGB
2F42Left hippocampal atrophySPS, CPSMultifocal (both arms), sporadic350 mg (4.32 mg/day)CBZ 1,800Myoclonus unchanged with PGB ↑ to 500 mg/day
3F24CryptogenicSPS, CPS, GTCSFocal (left arm), sporadic250 mg (3.79 mg/day)CBZ 2,400 CLB 10Myoclonus unchanged with PGB ↑ to 600 mg/day
4M23CryptogenicSPS, CPS, GTCSMultifocal (both arms + legs), sporadic50 mg (0.62 mg/day)CBZ 1,600 LTG 100Myoclonus unchanged with PGB ↑ to 200 mg/day

Myoclonus appeared at PGB doses of 50–600 mg/day (0.6–9.0 mg/kg per day). The duration of PGB therapy before onset of myoclonus ranged between 1 and 45 days. The highest frequency of myoclonic jerks (>20 per day) was observed under rapid titration to 600 mg PGB/day (patient 1, double-blind study), persisted at this dose, and decreased in frequency and intensity when the dose was reduced from 600 to 450 mg PGB/day during transition to the open add-on phase. The patient then reported only sporadic inner “jerks” without visible movements. The situation remained stable even after a later, more gradual increase of PGB to 500 mg/day. Because of a loss of the anticonvulsant effect after ∼16 months of treatment, the patient was taken off PGB medication, and the myoclonus totally disappeared.

In the other patients, the myoclonus was only subtle and did not significantly interfere with daily activities. According to patients and physician, a dose reduction of PGB was not necessary; with regard to the positive anticonvulsant effect of PGB; the medication was continued for 8–9 months until now. The patients continued to have sporadic (<10/day) myoclonic jerks with no apparent change in frequency or intensity over time, even when later the PGB dose was gradually increased.

The PGB doses in patients who did not develop myoclonus also varied between 200 and 600 mg/day (2.8–10.2 mg/kg per day) with an average of 400 PGB mg/day (5.9 mg/kg per day).

In addition to PGB, all patients received other AEDs, the doses of which were kept constant from ≥1 month before study entry onward. For the four patients with myoclonus, this medication consisted of CBZ, 1,600–2,400 mg/day, either alone or in combination with LTG, 100 mg/day, phenobarbital (PB),125 mg/day, or clobazam (CLB),10 mg/day, respectively. The 15 patients who did not develop myoclonus received CBZ (750–1,800 mg/day; eight patients), LTG (200–600; four), VPA (1,000–4,500; three), oxcarbazepine (OCBZ; 2,100–3,000; three), topiramate (TPM; 25–600; two), CLB (10–20; two), and/or PB (200; one). Apart from slightly lower CBZ doses in this group as compared with the myoclonus patients, there was no obvious difference with regard to the concomitant medication in the two groups.

While three patients developed multifocal myoclonus, one (patient 3) showed myoclonic jerks limited to the left arm (i.e., contralateral to the interictal and ictal EEG focus and in the same body region where paraesthesias and tonic contractions were observed during simple and complex partial seizures). However, with discordant findings in the presurgical evaluation and no pathology in MRI, the precise epileptogenic zone in this patient remained unclear.

In routine EEG recordings during wakefulness, no correlates of myoclonus were found in any study patient. Polygraphic studies including EMG-triggered back-averaging of the EEG in the patient with the highest frequency of myoclonic jerks (patient 1) also failed to show EEG changes. However, the number of myoclonic jerks recorded (∼25) was limited.


These data indicate a relatively high incidence (four of 19) of myoclonus associated with PGB therapy and confirm the observations of Asconapé et al. (1999), who reported two cases of new-onset myoclonus in six patients receiving PGB medication (12). Although the small size of the investigated population does not allow a true comparison, the potential of PGB to induce myoclonus seems to be at least as high as that described for the structurally similar anticonvulsant GBP. Recently Asconapé et al. (1) reported that myoclonus was observed in ∼12.5% of 104 patients treated with GBP. The authors explained the discrepancy to an incidence of only 0.1% during premarketing studies of GBP by the specific questioning of their patients. This may also account, at least in part, for the frequent observation of myoclonus in our patients.

There is evidence for a dose dependency because frequency of myoclonic jerks was highest in the highest PGB dose allowed in the study (600 mg/day) and decreased after dose reduction (patient 1). However, having once developed myoclonus, the frequency and intensity of myoclonic jerks in patients 2 through 4 did not seem to change when the PGB dose was further increased (Table 1). This indicates a threshold effect rather than a linear dose dependency.

Fifteen of 19 patients receiving PGB doses of 200–600 mg/day (average, 400 mg/day) did not develop myoclonus, which indicates that the occurrence of myoclonus does not depend on the PGB dose alone. Possible additional factors remain unclear, because the groups of patients with and without myoclonus in our studies showed no discernible difference with regard to their diagnoses, seizure types, or concomitant medications. Only the CBZ doses were slightly higher in the myoclonus group as compared with the other patients, and there was no patient with newly developed myoclonus who did not have a co-medication of CBZ. However, although worsening of epileptic myoclonus (13,14), or development of myoclonic jerking under CBZ therapy (5–7) have been reported, the occurrence of myoclonus in our patients is unlikely to be the exclusive result of concomitant CBZ therapy, because the patients had no myoclonus before starting PGB when the doses of concomitant AEDs (including CBZ) were kept constant for ≥1 month. Nevertheless, it is possible that myoclonus resulted from the particular combination of PGB and CBZ and would not have occurred with PGB in the absence of CBZ.

The mechanism by which PGB may induce myoclonus is hardly understood. Myoclonus is a well-known side effect of several AEDs (3,4,8,14,15). Especially GBP has been associated repeatedly with the occurrence of myoclonic jerks (1,2,15,16). At present, one can only speculate that similar to GBP, a lactam analogue may be involved in the induction of myoclonus by PGB (17).

In conclusion, myoclonus is a side effect of antiepileptic PGB therapy that seems to occur not infrequently. According to the results in our patients, its intensity is usually mild, and a discontinuation of therapy is not necessary. If severe, it appears to be reversible with a dose reduction of PGB.