Incidence of Dyskinesia in Patients with Early Parkinson’s Disease Treated with Ropinirole or Levodopa

Abstract & Commentary

Source: Rascol O, et al. A five-year study of the incidence of dyskinesia in patients with early Parkinson’s disease who were treated with ropinirole or levodopa. N Engl J Med 2000; 342:1484-1491.

Ropinirole, a d2 and d3 receptor nonergot agonist, reached the U.S. market in the fall of 1998. It has been found effective in ameliorating the motor manifestations of early and advanced Parkinson’s disease (PD) in several large clinical trials. In this double-blind, industry-sponsored study, PD patients were treated with ropinirole or levodopa and were followed for more than five years in order to determine the incidence of the development of dyskinesias. Two hundred sixty-eight patients with PD (Hoehn and Yahr stage 1-3) were enrolled. They were eligible for enrollment only if they required treatment with a dopaminergic agent, if they were not demented, if they had no history of psychosis, and if they had not received prior treatment for more than six weeks with either levodopa or any dopamine agonist.

The 268 patients were enrolled at 30 centers throughout Europe, Israel, and Canada. Two patients were assigned to the ropinirole group for each patient assigned to the levodopa arm. The study was performed in a double-blind, double-dummy fashion. After an initial dose-escalation period, patients were maintained at a stable dose of study drug. For patients whose symptoms were not adequately controlled by study drug alone, supplementary levodopa could be given in open-label fashion. The incidence and severity of dyskinesias were measured by the Unified Parkinson’s Disease Rating Scale, which asks patients to assess the severity and daily duration of dyskinesias.

Of the 268 patients who enrolled in the trial, 179 were assigned to the ropinirole arm and 89 to levodopa. The average daily dose of ropinirole during the study was 16.5 mg, and the average dose of levodopa was 753 mg. Dyskinesias developed in 20% of patients in the ropinirole group and 45% of patients in the levodopa group. Eight percent of the patients in the ropinirole group had "disabling" dyskinesias, compared to 23% of the patients given levodopa. The hazard ratio for remaining free of dyskinesias for the ropinirole group compared to the levodopa group was 2.82.

Unfortunately, the decrease in dyskinesias was not achieved without a price. Patients who received levodopa had a statistically significant improvement from baseline in motor performance as measured on section II of the Unified Parkinson’s Disease Rating Scale. This was not true for patients who received ropinirole: 32% of patients receiving ropinirole experienced freezing of gait, compared to 25% receiving levodopa. The incidence of nausea, insomnia, and postural hypotension was similar between the two groups. However, patients who received ropinirole had increased incidences of somnolence (27.4% vs 19.1%), hallucinations (17.3% vs 5.6%), and leg edema (14% vs 5.6%).

The premature withdrawal rate from the study was high—53% in the ropinirole group and 49% in the levodopa group. Of 179 patients initially assigned to receive ropinirole, 92 (51%) required supplementation with levodopa during the course of the trial. Only 29 patients were able to remain on ropinirole as monotherapy for the full five years.


Over the last decade, a shift in practice has occurred among movement disorder neurologists in how they treat patients with early PD. Most neurologists try to manage these patients with dopamine agonists first, instead of using levodopa. This reflects the commonly held belief that early treatment with levodopa accelerates the onset of motor fluctuations, including the wearing-off phenomenon and the development of dyskinesias. There are also theoretical reasons to support treatment with agonists. In vitro studies have shown that dopamine agonists may be neuroprotective, whereas levodopa has been shown to contribute to oxidative injury in cell culture models.

The present study confirmed the prediction of movement disorder neurologists, that treatment with dopamine agonists produces fewer dyskinesias than treatment with levodopa. The results of the study were heavily publicized when they were first announced in the summer of 1999. However, careful review of the design and analysis of the trial reveals several problems.

The study was designed to examine the effect of agonist monotherapy on "early" PD. To most clinicians, patients with "early" Parkinson’s have only limited functional disability. One-third of the patients enrolled in this trial, however, started with a Hoehn and Yahr stage of 2.5 (mild to moderate balance impairment) or 3.0 (no recovery on pull test). Most movement disorder neurologists would not consider a patient with any degree of balance impairment to be "early." At our center, balance impairment is a clear indication for starting levodopa.

The primary outcome measure of the ropinirole study was the development of dyskinesias. Rating dyskinesias is notoriously difficult and the validity of self-rated dyskinesias is unknown. Nevertheless, a dyskinesia rating of 1+ on item 32 of the Unified Parkinson’s Disease Rating Scale denotes dyskinesias of less than 25% of the waking day. A self-rating of 1+ on item 33 denotes dyskinesias that are only mildly disabling. Clinical experience suggests that involuntary movements of only 1+ severity are not troublesome. In fact, patients are much more bothered by slowness or stiffness than by occasional involuntary movements. Thus, the primary outcome measure of this study, the development of dyskinesias of 1+ severity on items 32 and 33 of the Unified Parkinson’s Disease Rating Scale, may not reflect patients’ perception of their disability.

Compared to levodopa, patients who received ropinirole failed to experience an equivalent symptomatic benefit. They enjoyed less improvement in their motor scores and had an earlier onset of gait freezing, a potentially serious symptom due to the increased risk of falling. Further, only 29 of 179 patients initially assigned to receive ropinirole were able to continue taking it as monotherapy after five years. This implies that ropinirole does not provide adequate, long-term symptomatic treatment for the majority of Parkinson’s patients who require dopaminergic therapy.

The incidence of adverse events in this trial was not significantly different between its two arms. There was, however, a clear trend toward increased somnolence, hallucinations, and lower extremity edema in the ropinirole group. Hallucinations were particularly troubling, occurring in 17.3% of the ropinirole population and only 5.6% of the levodopa group. Rascol and colleagues also state that there were no reports of falling asleep suddenly in either group. It is not known whether patients were specifically queried on this issue. Recent analyses have shown that patients under-report these events, and that they are often not captured by the standard adverse event questionnaires used in clinical trials.

How should neurologists interpret this study, and what effect will it have on their treatment of patients with PD? This trial showed that some patients can be effectively managed with ropinirole monotherapy for up to five years. However, most patients require supplemental levodopa to prevent unacceptable functional disability. At our center, we hesitate to use dopamine agonists in Parkinson’s patients older than the age of 70, as the incidence of adverse events (namely, somnolence and hallucinations) increases with age. Ropinirole monotherapy is a reasonable option in younger patients, particularly those without balance impairment and those who are in the early stages of the illness. Neurologists should keep in mind the recent reports of sudden sleep episodes associated with the newer dopamine agonists, and counsel appropriate patients with respect to safe driving. —steven frucht, md