Clozapine for Dyskinesias in Parkinson’s Disease

Abstract & Commentary

Source: Durif F, et al. Clozapine improves dyskinesias in Parkinson's disease: A double-blind, placebo-controlled study. Neurology. 2004;62:381-388.

Levodopa-induced dyskinesias are one of the most common motor complications of Parkinson’s disease (PD), typically occurring after 5-7 years of illness. These involuntary movements often affect young patients, and they can be disabling. Dyskinesias typically activate when patients use their limbs; they often interfere with basic activities of daily living, and can even be dangerous—for example, choreic dyskinesias of the neck may accelerate cervical spondylosis and even produce cord compression.

Although there are no approved medications to treat levodopa-induced dyskinesias, several strategies are commonly used. Reducing the interval and amount of each dose of levodopa, maximizing the dose of dopamine agonists, and even liquefying the daily levodopa and giving it in small hourly increments are useful in selected patients. One in 3 PD patients will derive benefit in dyskinesia control from treatment with amantadine. However, psychosis and leg edema often limit the usefulness of this drug. For patients with severe motor fluctuations and dyskinesias, bilateral subthalamic nucleus stimulation offers the only definitive treatment, but the procedure requires a patient who is a good surgical candidate and accepts the operative risks.

Recent interest has focused on clozapine, the gold-standard atypical neuroleptic, as a treatment for levodopa-induced dyskinesias. The drug possesses antagonist activity at the D1 and D2 receptors and agonist activity at striatal 5HT1 receptors. The present study evaluates clozapine in a double-blind, placebo-controlled trial as a treatment for levodopa-induced dyskinesias.

Fifty nondemented PD patients with disabling dyskinesias (defined by the commonly used rating items of the Unified Parkinson Disease Rating Scale) were randomized to receive either clozapine or placebo for a period of 8 weeks. The dose of clozapine was titrated according to the clinician and patient’s judgment, up to a maximum of 75 mg/d. Patients with dyskinesias were evaluated prior to enrollment and throughout the trial using a clinical rating scale to measure with dyskinesias and time "off." Patients were also asked to rate the severity of their dyskinesias every 2 weeks and were videotaped so that the severity of levodopa-induced dyskinesias could be rated in blinded fashion. In addition, patients were given an oral levodopa challenge at the beginning and end of the study, and the severity of dyskinesias were rated every 5 minutes for the duration of action of the challenge dose.

At the completion of the study, the mean dose of clozapine was 39 mg/d. There was a significant reduction in the duration of "on" periods with dyskinesias in the clozapine group: 5.7 hours per day vs 4.0 hours per day (P = .003 relative to the placebo group). There was a significant reduction in dyskinesias as the clozapine dose was increased, with benefit observed by the fourth week of the trial. No increase in "off" time or other deterioration in parkinsonism was observed. Levodopa-induced dyskinesias (as rated by the blinded videotaped analysis) following the levodopa challenge were also improved when patients were examined at rest. Activation-induced dyskinesias were not significantly different between the 2 groups. Somnolence and hypersalivation were more common in the clozapine group. Three patients taking clozapine developed hypereosinophilia, which resolved when the drug was discontinued.


This is one of the first trials to use a combination of patient self-rating, blinded videotape review, and levodopa challenge to measure the severity of dyskinesias. Durif and associates demonstrated that clozapine did not worsen parkinsonism and that it decreased the severity of dyskinesias. The drug was reasonably well tolerated for the duration of the trial.

The question facing neurologists in practice is whether clozapine should be used in patients with PD who develop significant levodopa-induced dyskinesias. Arguments against the use of clozapine include the trouble and expense of subjecting patients to a weekly blood draw for 6 months (followed by biweekly blood draw) to monitor the white blood cell count, the rare risk of myocarditis, the recently identified risk of inducing diabetes, and the known side effects of sedation, hypersalivation, and lowered seizure threshold that accompany the drug’s use.

In my opinion, the risks inherent with clozapine preclude its use as an antidyskinetic agent unless dyskinesias are disabling. They do not improve with standard anti-PD management strategies, and the patient obtains no benefit from amantadine. It would appear reasonable to consider treating this challenging group of patients with clozapine prior to consideration of deep-brain stimulation. — Steven Frucht, MD, Assistant Professor of Neurology, Movement Disorders Division, Columbia-Presbyterian Medical Center and Assistant Editor of Neurology Alert