Evidence Based Review Article

The Journal of Informed Pharmacotherapy 2001;4:221-223.

Is the dopamine agonist pramipexole an alternative to levodopa for treatment of early Parkinson disease?

Download a PDF version of this article

Reviewer: Debbie L. MacLeod, B.Sc. (Pharm); Priti Flanagan, Pharm.D.
Reviewer's email address:
Reviewer's profession/specialty:
Resident/Family practice and Geriatrics

Original Citation

Anon.  Pramipexole vs. levodopa as initial treatment for Parkinson disease: a randomized controlled trial. JAMA 2000;284:1931-1938. PubMed Cit

Overall Study Question

The objective of this study was to compare the development of dopaminergic motor complications after initial treatment of Parkinson disease (PD) with pramipexole versus levodopa.  Patients with early PD who required dopaminergic therapy were enrolled between October 1996 and August 1997.  Primary outcome was the time to first occurrence of any of three dopaminergic complications: wearing off, dyskinesias or on-off motor fluctuations.  Secondary outcomes included: changes in scores on UPDRS (Unified Parkinson's disease rating scale), PDQUALIF (Parkinson's disease quality of life scale) and EuroQol assessed from baseline to study completion.  A substudy to evaluate the potential neuroprotective effect of pramipexole on dopamine neurons was also performed. This was done by measuring decline in beta-CIT uptake using single photon emission computed tomography (SPECT).  b-CIT uptake is a marker of dopamine transporter density in dopaminergic neuron terminals in the striatum.  Postmortem data have shown reduced dopaminergic transporter density in PD patients.

Are the Results of the Study Valid?

1. Was assignment of patients randomized?

Yes.  Eligible patients were randomized 1:1 to either pramipexole or carbidopa/levodopa. Randomization was stratified by investigator and block.  

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Yes.  A flow chart was provided to graphically account for all those screened for study participation.  The 301 patients enrolled were randomized to two groups: 151 received pramipexole (23 withdrew due to adverse effects) and 150 received levodopa (19 withdrew due to adverse effects).  No patients were lost to follow-up.  Data analysis was based on an intention-to-treat basis and involved all randomized patients.

3. Were patients, their clinicians, and study personnel 'blind' to treatment?

Yes.  Both clinicians and patients were blinded to treatment assignment. This was accomplished using a double dummy technique.

4. Were the groups similar at the start of the trial?

Yes.  The two treatment groups were similar at baseline with regard to demographic and clinical variables.  All patients were in Hoehn and Yahr stages I, II or III, with similar divisions among these stages between the two treatment groups.  Hoehn and Yahr staging is a measure of PD severity typically used in clinical practice.  The pre-study quality of life assessment scores (PDQUALIF, EuroQol) were also similar between the two groups, as was prior use of levodopa, selegiline, amantadine and anticholinergic drugs.

5. Aside from the experimental intervention, were the groups treated equally?

Yes.  Patients in both groups received the same number of follow-up visits. For patients with residual disability, the dosage of either study drug was escalated during the first ten weeks, up to a maximum of 4.5mg pramipexole or 150/600mg carbidopa/levodopa.  From week 11 to month 23.5, investigators were permitted to add open label regular release levodopa (sustained release preparations were not allowed) to treat continuing or emerging disability in either group.

6. Overall, are the results of the study valid?


What were the Results?

1. How large was the treatment effect? 

The primary outcome was the time to first occurrence of any of three dopaminergic complications: wearing off, dyskinesias and on-off motor fluctuations.  Pramipexole treatment resulted in 42/151 patients (28%) developing dopaminergic complications vs. 76/150 patients (51%) treated with levodopa [hazard ratio (HR)=0.45, p<0.001].  These results translate into an absolute risk reduction (ARR) of 22.9% and a number needed to treat (NNT) of 4-5 patients (95% confidence interval (CI) = 3-8 patients).  Therefore, 4-5 patients would need to be treated over 23.5 months with pramipexole instead of levodopa to postpone the occurrence of a dopaminergic complication in one patient.  Differences between the two treatment arms were also reported for some of the secondary outcomes.  More patients in the pramipexole group (53%) versus the levodopa group (39%) required supplemental levodopa during the trial [HR=1.54 (1.09-2.17), p=0.02].

Mean improvement in the total UPDRS score from baseline to 23.5 months was greater in the levodopa group than in the pramipexole group 9.2 vs. 4.5 points (p<0.001).  Quality of life scores improved in both groups initially and then declined over time, with the levodopa group having higher scores at the end of the study (p=0.006).  Finally, somnolence was more common in pramipexole treated patients than in levodopa treated patients (32.4% vs. 17.3%, p=0.003).  In the substudy, there was a greater mean decline in striatal beta-CIT uptake in the levodopa group (24.8%) compared with the pramipexole group (20%).  This difference was not significant (p=0.15).

2. How precise was the estimate of the treatment effect?

Results associated with the primary endpoint included a HR of 0.45 with a 95% CI (0.30-0.66).  The 95%CI for ARR and NNT can be calculated to further demonstrate the precision of the results: ARR (15.2-33.5%), NNT (3-7).

Will the Results Help Me in Caring for My Patients?

1. Can the results be applied to my patient care?

Yes.  The well-defined inclusion/exclusion criteria would allow results to be applied to a definable population. However, the benefit in decreased dopaminergic complications demonstrated with pramipexole must be evaluated with consideration of the difference in efficacy, quality of life and adverse effects between levodopa and pramipexole. 

2. Were all clinically important outcomes considered?

Most outcomes were considered including time to development of dopaminergic complications, efficacy, adverse effects and quality of life.  Cost was not discussed.

3. Are the likely treatment benefits worth the potential harms and costs?

Although the time to development of dopaminergic complications was significantly greater in the pramipexole group, there were also significantly more patients who experienced somnolence, hallucinations and edema (generalized and peripheral).  As well, it is important to consider the difficulty that a patient with Parkinson disease may have in appreciating the benefit of delayed dopaminergic complications compared with the reality of the side effects and decreased quality of life (e.g. loss of a driver's license) that can be associated with the use of pramipexole.


Although levodopa is the gold standard for treatment of Parkinson disease, there has been considerable concern about the potential long term risks.  It is postulated that early exposure to levodopa in the course of Parkinson disease can lead to increased dopaminergic complications.  This has led to a search for alternative treatments that can be used to delay the initiation of levodopa treatment.  This study looks at the relationship between both levodopa and a dopamine agonist (pramipexole) to the development of dopaminergic complications and in a substudy, neuronal loss. The results do show a significant increase in time to onset of the dopaminergic complications with pramipexole; however, it is important to note that these dopaminergic complications still occurred with pramipexole.  Although, not statistically significant, pramipexole did show a slower decline in beta-CIT striatal uptake.  For those patients in whom the delay of initiation of levodopa is desirable, the use of pramipexole (a dopamine agonist) may be an alternative; however, the clinician must weigh the risks and benefits for each individual patient.

Copyright © 2000 by the Journal of Informed Pharmacotherapy. All rights reserved.