18

Dopamine Agonists

Mark A. Stacy
Barrow Neurological Institute,
Phoenix, Arizona, U.S.A.

INTRODUCTION
Dopamine agonists (DA) have been used to treat symptoms of Parkinson’s
disease (PD) since the late 1970s (1). These agents were initially introduced
to supplement the beneficial effect and possibly reduce the incidence of long-
term complications of levodopa. In the last 30 years, methodical
investigations of DA have demonstrated therapeutic benefit in all stages
of PD both in combination with levodopa and as monotherapy. More
recently, positron emission tomography (PET) and single photon emission
computed tomography (SPECT) imaging have demonstrated possible
benefit in patients randomized to a DA when compared to subjects
receiving levodopa (2–4). Increasingly, clinical, animal model, and cellular
data suggest not only a levodopa-sparing effect and a delay in the incidence
of motor fluctuations, but also a potential neuroprotective effect (5). A
number of hypotheses regarding this phenomenon have been proposed.
These include reduction of free radical formation by limiting levodopa
exposure or increase in the activity of radical-scavenging systems, perhaps
by changing mitochondrial membrane potential. In addition, some
investigators suggest that DA may enhance neurotrophic activity. However,

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 there is currently no evidence of neurological improvement clinically, using
trial designs with sufficient duration and washout periods to assess
neurological status beyond therapeutic response.
This chapter will review the history of DA usage in the treatment of
PD and provide a summary of data concerning efficacy, treatment
approaches, and comparison between commonly prescribed DA. In
addition, data suggesting long-term favorability when compared to
levodopa will be reviewed. Lastly, similarly designed clinical trials will be
discussed with direct comparative trials in an effort to better define the
relative efficacy of these agents.

DOPAMINE AGONISTS AND DOPAMINE RECEPTORS

The DA most often used in treating symptoms of PD include apomorphine,
bromocriptine, cabergoline, lisuride, pergolide, pramipexole, and ropinirole.
All of these agents activate D2 receptors, while pergolide has been shown to
be a mild D1 agonist, and pramipexole may have higher affinity for D3
(Table 1). Five subtypes of DA receptors have been identified and may be
classified into striatal (D1 and D2) receptors or cortical (D3, D4, and D5)
receptors. The D3–5 receptors are present in the limbic system and other
dopaminergic pathways. Although the different roles of D1 and D2
receptors in regulation of striatal function are more fully outlined elsewhere
in this volume, the D1 receptor (D1,5) family is associated with activation of
adenylate cyclase, and dopamine and DA activate the D2 receptor family
(D2–4) (6). Postmortem examination of brains of subjects with PD reveal
upregulation of striatal D2 and downregulation of the D1 receptors. It is
postulated that these changes lead to alteration of the indirect D2-mediated
pathway and disinhibition of the subthalamic nucleus. Intracortical
inhibition studies comparing apomorphine, a rapid-acting DA, to deep
brain stimulation found comparable changes in Unified Parkinson’s Disease
Rating Scale (UPDRS) and intracortical inhibition with bilateral sub-
thalamic nuclei or globus pallidus stimulation or with apomorphine
infusion, suggesting a connection between the nigral dopaminergic pathway
and the thalamo-cortical motor pathway (7).

Apomorphine
Because of the powerful emetic action of apomorphine, clinical usage of this
compound in treating PD has been avoided (8). More recently, this short-
acting DA has been developed as injectable and sublingual forms to be used
in ‘‘rescuing’’ PD patients from unpredictable off-periods. This therapy may

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 TABLE 1 Dopamine Agonists in Parkinson’s Disease
Dopamine
agonist D1 D2 D3 D4 D5 5-HT NE ACh

Dopamine þ þþ þþþ þþ þþþ 0 0 0

Bromocriptine
þþ þ þ þ
þ 0
Pergolide þ þþþ þþþ þþþ þ 0 þ 0
Pramipexole 0 þþþ þþþ þþ 0 þ þ þ
Ropinirole 0 þþþ þþþ 0 0 0 0 0

D ¼ dopamine; 5-HT ¼ 5 hydroxytryptamine; NE ¼ norepinephrine; ACh ¼ acetylcholine.

Source: Adapted from Ref. 6.

require initial treatment with domperidone, a peripheral DA receptor

blocking antiemetic agent. Dewey et al. (9) demonstrated a 62% improve-
ment in off-state UPDRS scores in subjects with advanced PD 20 minutes
after subcutaneously injecting apomorphine in a 2:1 randomized placebo-
controlled trial. This agent has been demonstrated to be effective as a
subcutaneously administered agent in 30 patients for up to 5 years of
therapy (10), and some follow-up studies of up to 8 years have demonstrated
long-term persistence of apomorphine efficacy. In a subset of patients who
could no longer tolerate subcutaneous injections, an intravenous (IV)
preparation is being evaluated. In one study of five subjects with severe
subcutaneous nodule formation who were followed for a mean of 7 months
(range 0.5–18 months) IV administered apomorphine appeared to produce
more consistent motor abilities, allowed for a reduction in oral medications
by an average of 59%, and decreased off time from 5.4 to 0.5 hours per day.
However, unanticipated intravascular thrombotic complications, secondary
to apomorphine crystal accumulation, were seen in two subjects (11).

Bromocriptine
Bromocriptine was first approved in the United States in 1978. This ergot
alkaloid is a partial D2 agonist and a mild adrenergic agonist. It also has
mild D1 and 5-hydroxytryptamine (5-HT) antagonist properties (Table 1).
When taken orally, bromocriptine is rapidly absorbed and 90% degraded
through first-pass hepatic metabolism. Peak drug levels are achieved in 70–
100 minutes, and it has a half-life of 3–8 hours. Less than 5% of the drug is
excreted into the urine, and bromocriptine is highly protein bound. The
drug is formulated into 2.5 mg scored tablets and 5 mg capsules (1). Dosing
titration usually begins at 1.25 mg/day and increases to a recommended

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 20 mg/day over the course of 7 weeks; however, successful treatment with
dosages higher than 60 mg/day has been reported (12) (Table 2).
While the side effects of the various DA are similar, only the ergot-
derived compounds have been associated with retroperitoneal fibrosis, a rare
but serious condition associated with severe pulmonary and renal
complications (13). Erythromelalgia, a painful discoloration of the shins,
may also be more prevalent in patients taking ergoline DA. The side effects
of nausea, vomiting, sleepiness, orthostatic hypotension, and hallucinations
are common to all DA, but in pivotal trials using bromocriptine they were
8–12% more common than in subjects receiving placebo (14) (Table 3).
Bromocriptine has been investigated extensively in de novo and
levodopa-treated populations. A recent systematic review of all randomized
controlled trials of bromocriptine monotherapy compared with levodopa
(LD) monotherapy in PD found that although numerous small trials have
been reported, methodological factors or lack of a control population has
led to a lack of evidence basis for clinical decisions (15–17). From 1974 to

TABLE 2 Agonist Titration Schedule

Time Bromocriptine Pergolide Pramipexole Ropinirole

Week 1 1.25 mg qd 0.05 mg qd 0.125 mg tid 0.25 mg tid

Week 2 1.25 mg bid 0.05 mg tid 0.25 mg tid 0.50 mg tid
Week 3 1.25 mg tid 0.10 mg tid 0.50 mg tid 0.75 mg tid
Week 4 2.50 mg tid 0.15 mg tid 0.75 mg tid 1.00 mg tid
Week 5 3.75 mg tid 0.25 mg tid 1.00 mg tid 2.00 mg tid
Week 6 5.00 mg tid 0.50 mg tid 3.00 mg tid
Week 7 5.00 mg qid 0.75 mg tid 4.00 mg tid
Week 8 1.00 mg tid
Maximum dosage 15.0 mg qid 2.00 mg qid 1.5 mg tid 8.00 mg tid

Source: Adapted Ref. 12.

TABLE 3 Dopamine Agonists in Parkinson’s Disease

Percentage
Dopamine agonist t1/2 (h) Metabolism N S H OH RPF

Bromocriptine 3–8 Hepatic 37 8 12 44 2–5

Pergolide 27 Hepatic 24 6 14 2 2–5
Pramipexole 8–12 Renal 18 13 19 16 0
Ropinirole 4–6 Hepatic 20 12 15 17 0

N ¼ nausea; S ¼ somnolence; H ¼ hallucinations; OH ¼ orthostatic hypotension;

RPF ¼ retroperitoneal or pulmonary fibrosis.
Source: Adapted from Ref. 1. Additional data from published package inserts.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 January 1999, six studies randomizing more than 850 patients to a
bromocriptine or a levodopa regimen are in the literature, but only two
trials were performed according to a double-blind design (16). These studies
indicate a reduced frequency of dyskinesias, and there was a trend toward
less wearing-off and fewer on-off problems in the bromocriptine group.
However, the statistically larger number of dropouts in the bromocriptine
group leaves these data subject to varied interpretations. In the treatment of
early PD bromocriptine may be beneficial in delaying motor complications
and dyskinesias with comparable effects on impairment and disability in
those patients who tolerate the drug.
Bromocriptine is well accepted as a treatment in advancing PD.
Although numerous studies have demonstrated this benefit in the past, a
more recent, well-designed, multicenter trial comparing bromocriptine to
placebo in patients with motor fluctuations has been reported. In this 9-
month study there was a 14% improvement in UPDRS activities of daily
living, and a 23.8% improvement in motor score. The bromocriptine arm
also demonstrated a 29.7% reduction in off-time while taking a mean dosage
of 22.8 mg/day (18).

Pergolide
Pergolide, an ergoline-derived DA, was approved for usage in the United
States in 1989. It is 10 times more potent than bromocriptine. Like
bromocriptine, pergolide has high affinity for the D2 receptor and mild a2-
adrenergic activity, but it does not have 5-HT activity. In addition, pergolide
has significant D3 activity and is the only DA with D1 agonist activity,
although this is mild (1) (Table 1). Pergolide is available in three tablet
sizes—0.05, 0.25, and 1.0 mg—and is usually titrated to an effective dosage
or an initial maximum dosage of 3 mg daily over the course of 6–8 weeks
(12) (Table 2). If clinical benefit is seen at 3 mg daily, this dosage may be
increased with disease progression to a typical maximum dosage of 6–8 mg/
day. Pergolide is rapidly absorbed from the gut and reaches a peak plasma
concentration in 1–3 hours. While the duration of action is typically 4–8
hours, the half-life ranges from 15 to 42 hours, with a mean of 27 hours.
Pergolide is highly protein bound, and >50% is excreted through the kidney
(1). Side effects associated with pergolide are similar to bromocriptine and
include retroperitoneal fibrosis, erythromelalgia, somnolence, orthostatic
hypotension, and hallucinations (14,19) (Table 3).
Pergolide has been demonstrated effective in both early- and late-stage
PD (20). In an open-label trial, Mizuno et al. demonstrated mild to
moderate to marked benefit in 47.5% of patients after 8 weeks of pergolide
at dosages up to 5 mg/day (21). Another open-label trial of 20 PD subjects

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 reported benefit in 7 subjects with a mean dosage of 0.85 mg/day. These
subjects were followed for 3 years. By 30 months, 7 subjects required
supplementation with levodopa, and by the end of the study the mean
dosage of pergolide had increased to 2.15 mg/day (19). A more recent de
novo PD study (PELMOPET-trial), with randomization to levodopa or
pergolide and concurrently evaluated using PET scanning, suggests a more
robust therapeutic effect. In this double-blind study, 294 subjects were
randomized to pergolide (n ¼ 148) or levodopa (n ¼ 146) and treated without
levodopa rescue for 36 months (4). An early report indicated that 77 subject
(52%) receiving pergolide compared to 90 subjects (61.6%) treated with
levodopa completed the study. Mean dosages for pergolide were 3.23 mg/
day and for levodopa were 504 mg/day. Although no statistical significance
was seen in study completers, differences were noted in change from baseline
UPDRS motor score (13.4 + 8.8 pergolide vs. 18.1 + 10.1 levodopa). As
expected, dyskinesias were three times more frequent and motor complica-
tions more severe in the levodopa group as captured by the UPDRS, part
IV. In addition, 88 subjects were followed by 18F-Dopa PET scans. Early
reports of these results show a decrease in uptake in the putamen of 7.9% in
the pergolide group and 14.5% in the levodopa group, but these differences
were nonsignificant (p ¼ 0.288) (Table 4). Evidence-based treatment data for
pergolide therapy in patients with motor fluctuations on levodopa are
available. In general, numerous small trials found data similar to those
reported by Olanow et al. (22). In this report of a 24-week, double-blind trial
of 377 subjects randomized to pergolide (189) or placebo (187), significant
improvements were seen in motor scores and off-time, and levodopa
dosages were reduced by approximately 25% in the pergolide group
(Table 5).
A review comparing efficacy data in adjunct therapy trials of pergolide
and bromocriptine found that pergolide was superior to bromocriptine in

TABLE 4 Relative Potencies of Dopamine Agonists in Early Parkinson’s Disease

UPDRS III
Effective dose Relative Relative dose Benefit vs. UPDRS III
(mg/d) potency range levodopa monoRx
Bromocriptine 15.–60 1:1 15–60 n/a þ 0.2
Pergolide 1.5–8 10:1 15–80 þ 3.0 vs
2.5a n/a
Pramipexole 1.5–4.5 10:1 15–45
3.4 vs
7.3b
6.0
Ropinirole 9.–24 2:1 18–48
0.8 vs
4.8c
4.5
a
36-month study.
b
23-month study.
c
40-month study.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 TABLE 5 Relative Potencies of Dopamine Agonists in Advanced Parkinson’s
Disease
Effective dose Off-time UPDRS III
(mg/d) LD; (mg/d) decrease (%) Benefit

Bromocriptine 22.6 ns-30% 0–30
3.0 (24%)

Pergolide 2.9 235 (20%) 32 n/a (35%)
Pramipexole 3.4 (22%) 15
4.0 (35%)
Ropinirole 7.5–24 191 (25%) 7 n/a

regard to improvements in motor function and activities of daily living (23).

In addition, more patients reported a ‘‘marked’’ or ‘‘moderate improve-
ment’’ with pergolide than with bromocriptine. However, no significant
difference in motor fluctuations, dyskinesias, levodopa dose reduction,
dropouts, or adverse events was found.

Pramipexole
Pramipexole was approved for use in the United States in July 1997. It is a
synthetic, nonergot DA. Like the ergot-derived DA, this agent is active at
the D2, D3, and D4 receptors. Pramipexole also has affinity for a- and b-
adrenoreceptors, acetylcholine receptors, and 5-HT receptors (Table 1). The
drug is available in 0.125, 0.25, 0.5, 1.0, and 1.5 mg tablets, and the usual
dosage is 3.0 mg/day over 5 weeks (12) (Table 2). When ingested, this drug
reaches peak plasma levels within 1–3 hours and has an elimination half-life
of 8–12 hours. The agent is excreted mostly unchanged in the urine, and
<20% of pramipexole is protein bound. Pivotal trials with pramipexole
report nausea, vomiting, somnolence, and orthostatic hypotension 0–13%,
higher than in subjects randomized to placebo (14) (Table 3). A condition
described as an ‘‘unexpected sleep episode’’ has been described and has also
been seen with other dopamine agents (24,25). In addition, pathological
gambling has been reported (26,27).
Pramipexole has been thoroughly studied in de novo and adjunctive
populations. Three major trials have evaluated the effectiveness of
pramipexole as monotherapy in early PD (28–31). A large dose ranging
trial (n ¼ 264) conducted by the Parkinson Study Group found that most
patients tolerated dosages of 6 mg or less of pramipexole. In this 10-week
study, 98% (placebo), 81% (1.5 mg/day), 92% (3.0 mg/day), 78% (4.5 mg/
day) and 67% (6.0 mg/day) of subjects tolerated drug to study conclusion
(28). A 20% benefit in motor score was seen in all active treatment groups,
and it was determined that the optimum dosage range was 1.5–4.5 mg/day.
In a 6-month study, 335 subjects were randomized to pramipexole (n ¼ 164)

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 or placebo (n ¼ 171). Investigators reported a greater than 20% improve-
ment in the UPDRS activities of daily living and motor scores in the active
treatment group (29). Side effects in these investigations included nausea,
somnolence, dizziness, and hallucinations (Table 4).
The Parkinson Study Group reported data comparing pramipexole to
levodopa in early PD (CALM-PD) (2,29,30). In this trial, 301 subjects were
randomized to pramipexole or placebo and were followed for 4 years. At the
conclusion of the trial, 52% of pramipexole and 74% of levodopa subjects
reached the primary endpoint of motor complications. Furthermore,
dyskinesias (25% vs. 54%) and wearing off (47% vs. 67%) were present in
a greater percentage of patients initiated with levodopa. However, UPDRS
assessments found significantly greater improvements (approximately 4
points) in subjects receiving levodopa (30). Eighty-two subjects in the
CALM-PD cohort also underwent sequential SPECT imaging with b-CIT
to assess striatal uptake of this dopamine transporter molecule (2).
Comparisons between the pramipexole (n ¼ 42) and the levodopa (n ¼ 40)
groups found statistically significant differences ranging from 6.4 to 9.5% in
transporter uptake at 22, 34, and 46 months, suggesting less decline in the
subjects receiving the dopamine agonist (Table 4). This study differs from
the PELMOPET study comparing pergolide to levodopa in that supple-
mental levodopa was allowed at the discretion of the investigator in either
group. The dosages of pramipexole averaged 2.78 mg/day, with 48% of
subjects receiving a levodopa supplement of a mean dosage of 264 mg/day,
while 36% of the levodopa-treated subjects required levodopa supplementa-
tion with a mean dosage of 509 mg/day.
Two large, randomized clinical trials using pramipexole in levodopa-
treated patients have demonstrated significant benefit in off time (31% and
15%), activities of daily living (22% and 27%), motor scores (25% and 35%),
and levodopa dosage reduction (27%) (18,31). The mean dosage of
pramipexole was 3.36 mg/day, and side effects including dyskinesias,
orthostatic hypotension, dizziness, insomnia, hallucinations, nausea, con-
fusion, and headache were seen (18,31) (Table 3).

Ropinirole
Ropinirole was approved in the United States in September 1997. This DA
is a nonergot compound with affinity for the D2 family of receptors, but not
the D1 or D5 receptors. In addition, unlike pergolide or pramipexole,
ropinirole lacks affinity for adrenergic, cholinergic, or serotonergic receptors
(Table 1). This drug is also rapidly absorbed from the gut with peak plasma
concentrations occurring in 1–2 hours and is 40% protein bound (1). The
elimination half-life is 6 hours, and the P450 CYP1A2 hepatic enzyme

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 pathway metabolizes the drug. Because of this, patients given ciprofloxacin
may have an increase in serum ropinirole concentrations. Because of the
difference in potency, when compared to pramipexole and pergolide, this
agent is often underdosed. Ropinirole is available in 0.25, 0.5, 1.0, 2.0, 3.0,
4.0, and 5.0 mg tablets and is usually titrated to 12 mg/day over a 7-week
period. Clinical improvement is usually not seen until patients are taking
6 mg/day (12) (Table 2). Side effects are similar to those seen with the other
DA and include nausea, somnolence, hallucinations, and orthostatic
hypotension (Table 3).
Ropinirole has been tested extensively in early and advanced PD
populations, and its development closely mirrors that of pramipexole. Adler
et al. randomized 241 untreated subjects to ropinirole (n ¼ 116) or placebo
(n ¼ 125) arms in a 24-week trial (32). Responders were defined as subjects
achieving at least a 30% improvement in UPDRS total and motor scores,
the primary outcome variables. In addition, subjects were assessed for time
to levodopa initiation and with a clinical global improvement scale. With an
average dosage of 15.7 mg daily, 47% of ropinirole subjects were identified
as responders, while only 20% of subjects responded in the placebo arm. The
mean changes in UPDRS of 24% in the ropinirole group and 3% in the
placebo group reached statistical significance. Time to levodopa significantly
differed, with 11% of ropinirole subjects and 29% of placebo subjects
requiring additional therapy. Requirement for levodopa therapy for the
ropinirole subjects was 16, 27, and 40% at 1, 2, and 3 years, respectively (1).
Studies have also been completed in de novo patients randomized to
ropinirole or placebo (ropinirole 056 study) and followed by PET scanning
(REAL-PET) (3,33). The 5-year clinical trial randomized 268 subjects to
ropinirole (n ¼ 179) or placebo (n ¼ 89) in a 2:1 fashion, allowing add-on
levodopa at the discretion of the investigator. Eighty-five subjects in the
ropinirole group (47%) and 45 subjects in the levodopa group (51%)
completed the study. In the ropinirole group 29 of the 85 patients (34%)
received no levodopa supplementation at 5 years. The analysis of the time to
dyskinesia showed a significant difference in favor of ropinirole, and at 5
years the cumulative incidence of dyskinesias, regardless of levodopa
supplementation, was 20% in the ropinirole group and 45% in the levodopa
group. The mean daily dose by the end of the study was 16.5 mg of
ropinirole. The average dosage for levodopa supplementation was 427 mg/
day. The subjects randomized to levodopa received an average of 753 mg/
day. Recently, the results of a PET analysis comparing ropinirole to
levodopa have been reported (REAL-PET) (3). This 2-year randomized trial
found a statistically significant difference in striatal uptake when comparing
ropinirole 13% to levodopa 20%. Clinical evaluation of these subjects found
that 14% of the ropinirole group vs. 8% of the levodopa group required

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 levodopa, while only 3% of subjects on ropinirole and 27% on levodopa
developed dyskinesias. Lang et al. (34) reviewed ropinirole data to assess
whether the development of dyskinesias in subjects exposed to early DA
monotherapy would have sustained benefit compared to levodopa, a
gradual return to the dyskinesia of those receiving early levodopa, or a rapid
return to the dyskinesias of levodopa subjects (34). In this review, no
differences were seen between groups, and a parallel rate of sustained
efficacy without dyskinesia was seen. A 6-month, placebo-controlled trial of
149 subjects with advanced PD randomized to ropinirole (n ¼ 95) or placebo
(n ¼ 54) followed for levodopa dosage and off-time reduction was
conducted. In this study, levodopa dosage was reduced an average of 31%
in ropinirole subjects compared to 6% in placebo subjects. Off-time was
reduced by 12% in the ropinirole group and 5% in the placebo group, a
difference in off-time of slightly over one hour per day (36).

Other Dopamine Agonists

Cabergoline is a once-daily, ergot-derived, dopamine agonist available in the
United States for the treatment of hyperprolactinemia. It has been
demonstrated to be effective in PD, but is prohibitively expensive in the
United States. Cabergoline has been evaluated in de novo and advanced PD
populations. In a double-blind, 2:1 placebo-controlled trial of 188 subjects
taking levodopa, the addition of cabergoline allowed for an 18% reduction
in levodopa, with a 16% improvement in motor scores (37). Clarke and
Deane have compared cabergoline to bromocriptine in a meta-analysis of
five randomized, double-blind, parallel group studies in 1071 patients (35).
Cabergoline produced benefits similar to bromocriptine in off-time
reduction, motor impairment and disability ratings, and levodopa dose
reduction over the first 3 months of therapy. Dyskinesia and confusion were
increased with cabergoline, but otherwise the frequency of adverse events
and withdrawals from treatment were similar with the two agonists.
Rotigotine is a novel DA that is unique in that it is delivered through a
skin patch transdermal system. In one multicenter phase II b trial, 316
patients were randomized to placebo or active drug in a 1:4 ratio and
followed by UPDRS for 4 weeks (38). In this dose-finding study, statistical
improvement in UPDRS activity of daily living and motor scores were seen
at 9, 13.5, and 18 mg/day. No changes were seen with placebo or at 4.5 mg/
day. Positive responder rates (>20% UPDRS change) increased with
dosage; responder rates were as follows: placebo, 29%; 4.5 mg, 38%; 9 mg,
45%; 13.5 mg, 57%; 18 mg, 53% (daily dosages) (38).

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 COMPARISONS BETWEEN DOPAMINE AGONISTS
At this time 15 comparative trials between DA are known: Tan and
Jankovic have summarized these studies and report conversion factors of
10:1 for bromocriptine to pergolide, 1:1 for pergolide to pramipexole, 1:6 for
pergolide to ropinirole, and 10:6 for bromocriptine to ropinirole (6). Hanna
et al. followed 21 stable subjects on pergolide switched safely to pramipexole
in a 1:1 ratio (39). Although not statistically significant, levodopa dosages
were reduced by 16.5%, and 13 of the 21 (62%) subjects reported
improvement with the change in regimen. Hauser et al. reported conversion
of stable subjects on levodopa and pramipexole to levodopa and ropinirole
in a 1:3 mg ratio. A gradual transition was somewhat better tolerated when
compared to rapid change (40). However, the difficulties reported by
subjects in retrospect may have been improved with a higher conversion
factor for ropinirole.
Although there are obvious difficulties when making direct compar-
isons in studies to determine dosage equivalence, a reasonable equation of
relative DA potency would suggest bromocriptine (6 10) ¼
pergolide ¼ pramipexole ¼ ropinirole (66) on a milligram-to-milligram
basis. Using the least potent and longest prescribed DA, a dosing
equivalence range may provide a useful measure of the treatment dosing
spectrum using these agents (Table 4). Using these assumptions, dosing
ranges appear to be higher in the ergot-related compounds, suggesting a
longer treatment horizon. However, at higher prescribed amounts, the
dosing curves for these agents tend to become more level, and the linear
improvement seen at midrange dosages may not be present at higher doses.
Perhaps a better measure of treatment response is to review similar trials of
DA therapy. In the early PD population, UPDRS data in similar, placebo-
controlled studies using pramipexole and ropinirole found remarkably
similar benefit (Table 4). Another potential comparison perspective is to
evaluate trials comparing two active interventions with the DA and
levodopa. In the imaging trials for pergolide, pramipexole, and ropinirole,
subjects treated with the agonist demonstrated similar, but less, benefit than
those with levodopa. It is unfortunate that the levodopa supplementation in
the pramipexole and ropinirole trials limited long-term monotherapy
assessment, so extensive direct comparisons cannot be made, but in general
the three trials produced similar benefit when compared to levodopa
response.
Because the data in the PELMOPET, CALM-PD, and 056/REAL-
PET trials represent, perhaps, the most rigorous and careful information
gathered about these compounds, it is useful to compare them as closely as
possible (2–4,31,33) (Table 4). Clinically, subjects treated with pergolide at

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 an average dosage of 3.23 mg/day demonstrated a motor scale decline of 3
points over 36 months, while subjects randomized to levodopa still
demonstrated benefit of 2.5 points in the same time interval. Although the
DA/levodopa trials using pramipexole and ropinirole allowed for levodopa
supplementation and had different durations, it is interesting to note that,
like the 5.5 point difference seen in the PELMOPET trial, the differences
between pramipexole (
3.4) and levodopa (
7.3) at 23.5 months
(difference ¼ 3.9 points) and ropinirole (
0.8) and levodopa (
4.8) at 60
months (difference ¼ 4 points) are highly similar. Furthermore, the small
differences between the PELMOPET (5.5 UPDRS motor points) and the
CALM-PD (3.9 points) and 056 (4.0) trials may be explained by the
levodopa supplementation allowed in the latter studies. The functional
imaging data from these investigations are also similar for dosages of 3.23,
2.78, and 16.5 mg/day for pergolide, pramipexole, and ropinirole, respec-
tively (Fig. 1). F-dopa PET demonstrates ropinirole striatal decline 65.0% of
levodopa at 24 months and pergolide striatal decline 54.5% of levodopa at
36 months. SPECT imaging with b-CIT demonstrates 52.6%, 55.6%, and
62.7% pramipexole to levodopa decline at 22, 34, and 46 months of
treatment. The imaging impact of added levodopa in the pramipexole and
ropinirole groups is unknown.
In summary, similar designs between pergolide, pramipexole, and
ropinirole demonstrate similar benefits in terms of levodopa dosage
reduction, levodopa percent reduction, treatment responders, and decrease
in off-time in adjunctive therapy trials (Tables 3, 4). In these studies subject
selection, methodological design, and data collected differed to the point
that trends are less reliable than in the early patient studies, but in general
similar improvements in all variables were seen.

TREATMENT WITH DOPAMINE AGONISTS

Initiation of Therapy in a New-Onset PD Patient
DA provide substantial improvement in PD symptoms while delaying the
development of early morning foot dystonia, motor fluctuations, and
dyskinesias (41). In addition, similar trials comparing DA (pergolide,
pramipexole, ropinirole) to levodopa in a randomized fashion suggest
possible long-term benefit by functional imaging measures (42). In a clinical
setting of a 30-year-old patient, it is quite compelling to delay levodopa
therapy in favor of DA because of the potentially long clinical horizon (43).
Conversely, in an 80-year-old patient with other health concerns, treatment
with levodopa may be better tolerated. The decisions regarding initial
therapy in the 50 years between these two examples is dependent on the

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 FIGURE 1 Dopamine agonist imaging data. Striatal decline by F-dopa and b-CIT
in Parkinson’s disease.

health of the patient, the relationship of the physician and the patient, the
side effect profiles of the DA and levodopa, and, unfortunately, the cost of
the drug (Table 2).
Dopaminergic medications carry similar side effect profiles, including
nausea, sleepiness, confusion, orthostatic hypotension, and hallucinations
(Table 3). Besides these problems, lower extremity edema, hair loss, and
weight gain have also been seen with DA, and the ergoline derivatives
bromocriptine, cabergoline, and pergolide also carry a slight risk of
erythromelalgia (a reddish discoloration of the legs), and pulmonary and
retroperitoneal fibrosis has been reported in 2–5% of subjects exposed to
these agents (1). With the exception of nausea, DA are more likely than
levodopa to cause these clinical difficulties, and discussion of the potential
side effects at the time of prescribing will greatly aid in the tolerance of any
new pharmacological agent. Because the statistical spectrum of side effects
of these agents are quite similar but vary from patient to patient, it is
important for any patient to understand that if he or she does not tolerate
the first DA, there is no reason to expect that the other DA will not provide
benefit.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 Lastly, the nonergoline agonists pramipexole and ropinirole have been
associated with ‘‘unexpected sleep episodes,’’ and although this problem has
been reported with pergolide and levodopa, it remains difficult to refute the
observation that this symptom had not been reported prior to the use of
pramipexole and ropinirole (24,25). In general, patients reporting excessive
daytime sleepiness should be monitored closely with the increase of any
dopaminergic therapy, especially in the first 3 months after the change.
Initiation of DA therapy is somewhat dependent on the needs and
emotional state of the patient (44). Each of the DA requires a titration
period of 4–8 weeks (Table 3). In the healthy patient seeking to improve
quickly, initiation of a rapidly titratable agent may be preferred, while the
more slowly titrated schedules may suit the needs of a patient reluctant to
take any drugs. However, each patient should be reminded that the
differences in titration time usually are less than 3 weeks, a brief period of
time in the context of a 20-year treatment horizon (Table 2).

Initiation of Dopamine Agonists as Adjunctive Therapy

The initiation of therapy in the early patient is somewhat arbitrary with the
exception of pramipexole and renal metabolism vs. the other DA and
hepatic metabolism. With advancing PD, the addition of a DA should
minimize the risk of aggravating further symptoms of nausea, sleepiness,
orthostatic hypotension, and other problems (45) (Table 3).

CONCLUSIONS
The development of DA, particularly pergolide, pramipexole, and
ropinirole, has gradually shifted treatment paradigms in PD. In the last
20 years, many parkinsonologists have moved from using DA as adjunctive
therapy to levodopa to initiating antiparkinsonian therapy with one of these
agents in otherwise healthy subjects (41). More recently, imaging data with
SPECT and PET scanning have produced debate regarding the possible
‘‘neuroprotective’’ advantages of DA when compared to levodopa (2–4). In
this regard some have questioned whether these agents should be initiated
sooner in the disease course, perhaps before obvious disability develops.
Regardless of when DA therapy is initiated, each patient benefits from the
choice of several agents for treating the symptoms of PD, and it is the
responsibility of the physician to provide the information regarding the
reasons for using this class of drugs and for choosing one agent over
another.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 REFERENCES
1. Factor SA. Dopamine agonists. Med Clin North Am 1999; 83:415–443.
2. Parkinson Study Group. Dopamine transporter brain imaging to assess the
effects of pramipexole vs. levodopa on Parkinson disease progression. JAMA
2002; 287:1653–1661.
3. Whone AL, Remy P, Davis MR, et al. The REAL-PET study: slower
progression in early Parkinson’s disease treated with ropinirole compared with
1-dopa. Neurology 2002; 58(suppl 3): A82–83.
4. Oertel WH, Schwartz J, Leenders DL, et al. Results of a 3-year randomized,
double-blind, PET-controlled study of pergolide vs. levodopa as monotherapy
in Parkinson’s disease (PELMOPET trial). J Neurol Sci 2001;
187(suppl):S444.
5. Le WD, Jankovic J. Are dopamine receptor agonists neuroprotective in
Parkinson’s disease? Drugs Aging 2001; 18(6):389–396.
6. Tan E-K, Jankovic J. Choosing dopamine agonists in Parkinson’s disease.
Clin Neuropharmacol 2001; 24:247–253.
7. Pierantozzi M, Palmieri MG, Mazzone P, Marciani MG, Rossini PM, Stefani
A, Giacomini P, Peppe A, Stanzione P. Deep brain stimulation of both
subthalamic nucleus and internal globus pallidus restores intracortical
inhibition in Parkinson’s disease paralleling apomorphine effects: a paired
magnetic stimulation study. Clin Neurophysiol 2002; 113:108–113.
8. Poewe W, Wenning GK. Apomorphine: an underutilized therapy for
Parkinson’s disease. Mov Disord 2000; 15(5):789–794.
9. Dewey RB Jr, Hutton JT, LeWitt PA, Factor SA. A randomized, double-
blind, placebo-controlled trial of subcutaneously injected apomorphine for
parkinsonian off-state events. Arch Neurol 2001; 58:1385–1392.
10. Stocchi F, Vacca L, De Pandis MF, Barbato L, Valente M, Ruggieri S.
Subcutaneous continuous apomorphine infusion in fluctuating patients with
Parkinson’s disease: long-term results. Neurol Sci 2001; 22(1):93–94.
11. Manson AJ, Hanagasi H, Turner K, Patsalos PN, Carey P, Ratnaraj N, Lees
AJ. Intravenous apomorphine therapy in Parkinson’s disease: clinical and
pharmacokinetic observations. Brain 2001; 124(Pt 2):331–340.
12. Pahwa R, Lyons KE. Solving the puzzle of Parkinson’s therapy. Pract Neurol
2002; 1:17–18,22–24.
13. Goetz CG, Tanner CM, Glantz RH, Klawans HL. Chronic agonist therapy
for Parkinson’s disease: a 5-year study of bromocriptine and pergolide.
Neurology 1985; 85:749–751.
14. Physician’s Desk Reference, 56th ed. Montvale, NJ: Medical Economics Co.,
2002.
15. Factor SA, Weiner WJ. Viewpoint: Early combination therapy with
bromocriptine and levodopa in Parkinson’s disease. Mov Disord 1993;
8:257–262.
16. Clarke CE, Speller JM. Pergolide versus bromocriptine for levodopa-induced
motor complications in Parkinson’s disease. Cochrane Database Syst Rev
2000; (2):CD000236.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 17. Ramaker C, van Hilten JJ. Bromocriptine versus levodopa in early
Parkinson’s disease. Cochrane Database Syst Rev 2000; (3):CD002258.
18. Guttman M, International Pramipexole-Bromocriptine Study Group. Double-
blind comparison of pramipexole and bromocriptine treatment with placebo
in advanced Parkinson’s disease. Neurology 1997; 49:1060–1065.
19. Jankovic J. Long-term study or pergolide in Parkinson’s disease. Neurology
1985; 35:296–299.
20. Bonuccelli U, Colzi A, Del Dotto P. Pergolide in the treatment of patients
with early and advanced Parkinson’s disease. Clin Neuropharmacol 2002;
25:1–10.
21. Mizuno Y, Kondo T, Narabayashi H. Pergolide in the treatment of
Parkinson’s disease. Neurology 1995; 45(suppl):S13–21.
22. Olanow CW, Fahn S, Muenter M, et al. A Multicenter double-blind placebo
controlled trial of pergolide as adjunct to Sinemet in the treatment of
Parkinson’s disease. Mov Disord 1994; 9:40–47.
23. van Hilten JJ, Ramaker C, Van de Beek WJ, Finken MJ. Bromocriptine for
levodopa-induced motor complications in Parkinson’s disease. Cochrane
Database Syst Rev 2000; (2):CD001203.
24. Frucht S, Rogers JD, Greene PE, et al. Falling asleep at the wheel; motor
vehicle mishaps in persons taking pramipexole and ropinirole. Neurology
1999; 52:1908–1910.
25. Stacy MA. Sleep attacks: Is it the drug or the disease? Drugs Aging 2002;
19:733–739.
26. Samanta JES, Stacy M. Pathologic gambling in Parkinson’s disease. Mov
Disord 1998; 15(suppl 3):111.
27. Gschwandtner U, Astar J, Renard S, Fuhr P. Pathologic gambling in patients
with Parkinson’s disease. Clin Neuropharmacol 2001; 24:170–172.
28. Hubble JP, Koller WC, Cutler NR, et al. Pramipexole in patients with early
Parkinson’s disease. Clin Neuropharmacol 1995; 18:338–347.
29. Parkinson Study Group. Pramipexole vs. levodopa as and initial treatment for
Parkinson disease. JAMA 2000; 284:1931–1938.
30. Holloway RG and the Parkinson Study Group. Pramipexole versus levodopa
as initial treatment for Parkinson’s disease: a four-year randomized controlled
trial. Neurology 2002; 58(suppl 3):A81–82.
31. Lieberman AN, Ranhosky A, Korts D. Clinical evaluation of pramipexole in
advanced Parkinson’s disease: results of a randomized, placebo-controlled,
parallel group study. Neurology 1997; 49:162–168.
32. Adler CH, Sethi KD, Hauser RA, et al. Ropinirole for the treatment of early
Parkinson’s disease. Neurology 1997; 49:393–399.
33. Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, Lang AE. A
five-year study of the incidence of dyskinesia in patients with early Parkinson’s
disease who were treated with ropinirole or levodopa. 056 Study Group. N
Engl J Med 2000; 342:1484–1491.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.

 34. Lang AE, Rascol O, Brooks DJ, et al. The development of dyskinesias in
Parkinson’s disease patients receiving ropinirole and given supplemental
l-dopa. Neurology 2002; 58(suppl 3):A82.
35. Lieberman A, Olanow CW, Sethi K, et al. A multicenter trial of ropinirole as
adjunct treatment for Parkinson’s disease. Neurology 1998; 51:1057–1061.
36. Hutton JT, Koller WC, Ahlskog JE, et al. Multicenter, placebo-controlled
trial of cabergoline taken once daily in Parkinson’s disease. Neurology 1996;
46:1062–1065.
37. Clarke CE, Deane KD. Cabergoline versus bromocriptine for levodopa-
induced complications in Parkinson’s disease. Cochrane Database Syst Rev
2001; (1):CD001519.
38. Bianchine J, Poole K. Efficacy and dose response of the novel transdermally
applied dopamine agonist rotigotine CDS in early Parkinson’s disease.
Neurology 2002; 58(suppl 3):A162–163.
39. Hanna PA, Ratkos L, Ondo WG, et al. A comparison of the therapeutic
efficacy of pergolide and pramipexole in Parkinson’s disease. J Neural Transm
2001; 108:63–70.
40. Hauser R, Reider C, Stacy M, et al. Acute vs. gradual pramipexole to
ropinirole switch. Mov Disord 1998; 15(suppl):133.
41. Olanow CW, Koller WC. An algorithm for the management of Parkinson’s
disease. Neurology 1998; 50(suppl):S10–14.
42. Hubble JP. Long-term studies of the dopamine agonists. Neurology 2002;
58(suppl):S42–51.
43. Poewe W. Should treatment of Parkinson’s disease be started with a dopamine
agonist? Neurology 1998; 51(suppl):S21–24.
44. Hauser R, Zesiewicz TA. Management of early Parkinson’s disease. Med Clin
North Am 1999; 393–414.
45. Olanow CW. The role of dopamine agonists in the treatment of early
Parkinson’s disease. Neurology 2002; 59(suppl):S33–41.

Copyright 2003 by Marcel Dekker, Inc. All Rights Reserved.