Pramipexole

What it is

Pramipexole is a synthetic non-ergot dopamine agonist taken by mouth, supplied as pramipexole dihydrochloride immediate-release and extended-release tablets. It acts directly on dopamine receptors rather than being converted into dopamine, and it binds the D3 receptor subtype with higher affinity than D2 or D4. It is almost entirely cleared unchanged by the kidneys, so kidney function governs how much stays in the body.

What the research says

For Parkinson's disease, trials show it improves motor symptoms and, started before levodopa, delays motor complications, but controls symptoms less well than levodopa and causes more somnolence, oedema and hallucinations. For restless legs syndrome, 12-week placebo-controlled trials show a real reduction in symptom scores. A delayed-start trial found no sign that it slows the underlying disease. The harms that matter most are not rare: impulse-control disorders, sudden-onset sleep, hallucinations in older people, augmentation of restless legs on long-term use, and a withdrawal syndrome on stopping.

Evidence grade: Well established.

How it works

Drug class: Non-ergot dopamine agonist, selective for the D2 receptor subfamily and binding D3 with higher affinity than D2 or D4

Pramipexole binds and switches on dopamine receptors directly, standing in for the dopamine that is lost in Parkinson's disease. It prefers the D3 subtype. In Parkinson's disease the effect is thought to come from stimulating dopamine receptors in the striatum; for restless legs syndrome the label states the precise mechanism is not known. (Source 1)

What it is used for

• Starting treatment with pramipexole instead of levodopa left fewer people with wearing off, dyskinesia or on-off fluctuations at 23.5 months (28% versus 51%), but levodopa controlled symptoms better and pramipexole caused more somnolence. A Cochrane meta-analysis of all dopamine agonists found the same trade-off. Evidence: established. (Source 2)
• A 12-week placebo-controlled trial in 344 people found larger falls in the International RLS Study Group score and more clinician-rated improvement than placebo, though the placebo response was itself large and the three doses did not line up in order of benefit. Augmentation - symptoms worsening on treatment - is the long-term concern, and the only pramipexole-specific figure we could source for it is a single retrospective series of 103 patients at one hospital. Evidence: established. (Source 3)
• The 535-patient delayed-start PROUD trial found no difference in disability score or in dopamine-transporter imaging between starting pramipexole at once and starting it 6 to 9 months later. Evidence: not-supported. (Source 4)
• In the 39-participant PAX-BD trial the 12-week primary outcome was not statistically significant (p = 0.087), although some longer-term secondary outcomes favoured pramipexole and hypomanic symptoms increased. The trial closed early and its authors say no change in practice can be recommended. Evidence: limited. (Source 5)
• One 60-patient single-centre 14-week placebo-controlled trial reported a 36% versus 9% fall in a visual-analogue pain score and 42% versus 14% achieving at least a halving of pain. It has not been replicated at scale, and impulse-control problems have since been reported in people given pramipexole for fibromyalgia. Evidence: limited. (Source 6)

Interactions

• Dopamine antagonists (antipsychotics such as phenothiazines, butyrophenones and thioxanthenes; metoclopramide) (label): These drugs block the receptors pramipexole acts on, so they can cancel out its effect. (Source 7)
• Cimetidine and other drugs cleared by the kidney's cationic transporter (ranitidine, diltiazem, triamterene, verapamil, quinidine, quinine) (pharmacokinetic study): They compete with pramipexole for the same route out of the body, so blood levels rise. (Source 8)
• Alcohol (label): Alcohol adds to the sleepiness pramipexole causes and raises the chance of falling asleep when awake. (Source 9)
• Alcohol and other sedating medicines, in the context of sudden-onset sleep (label): The label tells prescribers to ask specifically about alcohol and sedatives before starting, because they raise the risk of daytime sleep attacks. (Source 10)
• Food (any meal) (pharmacokinetic study): A meal does not change how much pramipexole is absorbed; it delays the peak blood level by about an hour. (Source 8)
• Levodopa (carbidopa/levodopa) (pharmacokinetic study): Pramipexole does not change how much levodopa is absorbed overall, but raises its peak level and brings the peak forward. (Source 8)
• Dietary supplements (no documented pramipexole-specific interaction found) (label): We found no documented interaction between pramipexole and a named supplement. The only interaction the label's drug-interactions section carries is with dopamine antagonists, and the clinical-pharmacology section lists only prescription drugs sharing its renal transporter. Iron matters to restless legs syndrome itself but is not described as interacting with pramipexole. (Source 7)

Stopping it

• The label records a withdrawal syndrome on tapering or stopping dopamine agonists - apathy, anxiety, depression, fatigue, insomnia, sweating and pain - and states that these symptoms generally do not respond to levodopa. (Source 11)
• In a retrospective cohort, 5 of 26 people (19%) tapered off a dopamine agonist developed the syndrome; all of them had had a dopamine-agonist-related impulse control disorder beforehand, and they had taken higher doses for longer. (Source 12)
• A review of the syndrome puts the figure at up to 19% of Parkinson's patients who undergo a taper, names impulse control behaviour disorders and higher agonist dose as risk factors, and states there is no standard treatment, so recognising who is at risk before tapering is what matters. (Source 13)
• The extended-release label carries the same withdrawal warning, so this is not a property of one formulation. (Source 14)
• In restless legs syndrome the pattern is different again: in the trials pramipexole was stopped without a taper, and in a 26-week placebo-controlled trial people reported their restless legs symptoms were worse than their own untreated baseline when the drug was withdrawn suddenly - rebound, not the dopamine agonist withdrawal syndrome. (Source 15)

What goes wrong

In the largest cross-sectional study of its kind, people with Parkinson's disease taking a dopamine agonist had about 2.7 times the odds of an impulse control disorder as those not taking one (odds ratio 2.72, 95% CI 2.08-3.54). (Source 16)

• Survey study, Moderate certainty.
• Size: 3,090 patients with treated idiopathic Parkinson's disease at 46 movement disorder centres.
• Who: People receiving routine clinical care in the US and Canada; raters blinded to medication status.
• How long: point prevalence.
• Result: An impulse control disorder was found in 13.6% overall - gambling 5.0%, compulsive sexual behaviour 3.5%, compulsive buying 5.7%, binge-eating 4.3%, with 3.9% having two or more. Prevalence was 17.1% on a dopamine agonist versus 6.9% off one (odds ratio 2.72, 95% CI 2.08-3.54; P<.001) - an absolute difference of about 10 percentage points. Pramipexole and ropinirole did not differ (17.7% versus 15.5%; OR 1.22, 95% CI 0.94-1.57; P=.14).
• Funding: not stated in the abstract (clinicaltrials.gov NCT00617019)

Limit of this finding: This is a cross-sectional study, so it cannot establish that the drug caused the behaviour; the authors themselves list younger age, being unmarried, smoking and a family history of gambling problems as independently associated. A reader should not read the odds ratio as a risk the drug creates in any individual. Two things in the paper itself need care. The four individual behaviours (gambling 5.0%, compulsive sexual behaviour 3.5%, compulsive buying 5.7%, binge-eating 4.3%) add to 18.5%, more than the 13.6% overall figure, because 3.9% of people had two or more; they must not be added together. And the paper's conclusion describes the effect as '2- to 3.5-fold', which is the span of the confidence interval, not the measured odds ratio of 2.72.

Impulse control disorders were more common in patients treated with a dopamine agonist than in patients not taking a dopamine agonist (17.1% vs 6.9%; odds ratio [OR], 2.72; 95% confidence interval [CI], 2.08-3.54; P < .001). Impulse control disorder frequency was similar for pramipexole and ropinirole (17.7% vs 15.5%; OR, 1.22; 95% CI, 0.94-1.57; P = .14).

People taking pramipexole have fallen asleep without warning while driving, causing crashes, and the attacks stopped when the drug was withdrawn. (Source 17)

• Case series, Very low certainty.
• Size: 9 patients (8 on pramipexole, 1 on ropinirole)
• Who: People with Parkinson's disease taking a non-ergot dopamine agonist.
• How long: not stated.
• Result: Nine patients fell asleep while driving and caused accidents; five had no warning before falling asleep. The attacks ceased when the drugs were stopped. No rate and no comparison group - this is a case series, not an incidence estimate.
• Funding: not stated.

Limit of this finding: Nine reported cases with no comparison group and no denominator: this shows that sudden-onset sleep happens, not how often. No rate can be read from it.

The authors report a new side effect of the dopamine agonists pramipexole and ropinirole: sudden irresistible attacks of sleep. Eight PD patients taking pramipexole and one taking ropinirole fell asleep while driving, causing accidents. Five experienced no warning before falling asleep. The attacks ceased when the drugs were stopped.

Hallucinations were several times more common on pramipexole than on placebo in Parkinson's disease trials, and the excess was much larger in people over 65. (Source 18)

• Randomized trial, Moderate certainty.
• Size: 388 versus 235 (early Parkinson's disease); 260 versus 264 (advanced, on levodopa)
• Who: Participants in pooled double-blind placebo-controlled pre-marketing trials.
• How long: the double-blind trial periods.
• Result: Early Parkinson's disease: hallucinations in 9% (35 of 388) on pramipexole versus 2.6% (6 of 235) on placebo. Advanced disease with levodopa: 16.5% (43 of 260) versus 3.8% (10 of 264). Risk relative to placebo was 1.9 times greater under age 65 and 6.8 times greater over 65 in early disease; 3.5 and 5.2 times in advanced disease. Discontinuation for hallucinations: 3.1% and 2.7% versus about 0.4% on placebo.
• Funding: industry (manufacturer's label data)

Limit of this finding: Every figure here comes from Parkinson's disease trials. In the same section of the label the only restless legs figure is one patient out of 889, so these percentages must not be read as the risk for someone taking pramipexole for restless legs. The same section also states that people with a major psychotic disorder should ordinarily not be treated with a dopamine agonist, because of the risk of making the psychosis worse.

In the three double-blind, placebo-controlled trials in early Parkinson's disease, hallucinations were observed in 9% (35 of 388) of patients receiving pramipexole dihydrochloride tablets, compared with 2.6% (6 of 235) of patients receiving placebo. In the four double-blind, placebo-controlled trials in advanced Parkinson's disease, where patients received pramipexole dihydrochloride tablets and concomitant levodopa, hallucinations were observed in 16.5% (43 of 260) of patients receiving pramipexole dihydrochloride tablets compared with 3.8% (10 of 264) of patients receiving placebo. Hallucinations were of sufficient severity to cause discontinuation of treatment in 3.1% of the early Parkinson's disease patients and 2.7% of the advanced Parkinson's disease patients compared with about 0.4% of placebo patients in both populations.

In restless legs syndrome trials nausea and somnolence were the adverse reactions clearly more common on pramipexole than on placebo. (Source 19)

• Randomized trial, Moderate certainty.
• Size: 575 pramipexole versus 223 placebo across three double-blind trials.
• Who: People with restless legs syndrome.
• How long: up to 12 weeks.
• Result: Nausea 16% versus 5%, somnolence 6% versus 3%, constipation 4% versus 1%, diarrhoea 3% versus 1%, dry mouth 3% versus 1%, fatigue 9% versus 7%, headache 16% versus 15%. About 7% of 575 stopped for adverse reactions versus 5% of 223 on placebo; nausea was the commonest reason (1%).
• Funding: industry (manufacturer's label data)

The most common adverse reactions with pramipexole dihydrochloride tablets in the treatment of RLS (observed in >5% of pramipexole-treated patients and at a rate at least twice that observed in placebo-treated patients) were nausea and somnolence. Occurrences of nausea and somnolence in clinical trials were generally mild and transient. Approximately 7% of 575 patients treated with pramipexole dihydrochloride tablets during the double-blind periods of three placebo-controlled trials discontinued treatment due to adverse reactions compared to 5% of 223 patients who received placebo. The adverse reaction most commonly causing discontinuation of treatment was nausea (1%).

Nineteen percent of people tapered off a dopamine agonist developed a stereotyped withdrawal syndrome that did not respond to levodopa, and every affected person had a prior impulse control disorder. (Source 12)

• Cohort study, Low certainty.
• Size: 26 of 93 non-demented patients underwent a dopamine agonist taper.
• Who: Outpatients with Parkinson's disease at a tertiary movement disorders clinic.
• How long: retrospective, over the course of the cohort study.
• Result: 5 of 26 (19%) developed the syndrome. Affected people had higher baseline agonist use (mean 420 [SD 170] versus 230 ¹⁸⁰ dopamine-agonist levodopa equivalent daily doses; P = .04) and higher cumulative exposure (1800 ¹²⁰⁰ versus 700 ⁹⁰⁰ DA-LEDD-years; P = .03). Symptoms included anxiety, panic attacks, agoraphobia, depression, dysphoria, diaphoresis, fatigue, pain, orthostatic hypotension and drug cravings.
• Funding: not stated.

Limit of this finding: This is a small retrospective single-clinic series, not a population rate, and the denominator is 26 people. A reader should not treat 19% as the chance of withdrawal for any individual stopping pramipexole. The 19% is 5 of the 26 people in the clinic's records who actually tapered an agonist - not 19% of the 93-patient cohort and not 19% of the 40 who were on an agonist. Every characteristic described rests on those 5 people.

Of these 26 subjects, 5 (19%) developed DAWS and 21 (81%) did not. All subjects with DAWS had baseline DA-related impulse control disorders. Symptoms of DAWS resembled those of other drug withdrawal syndromes and included anxiety, panic attacks, agoraphobia, depression, dysphoria, diaphoresis, fatigue, pain, orthostatic hypotension, and drug cravings.

The only pramipexole-specific figure for augmentation - restless legs symptoms getting worse, and starting earlier in the day, on treatment - comes from a single retrospective review of 103 patients at one Chinese hospital, in which 15 were judged to have augmentation. (Source 20)

• Cohort study, Low certainty.
• Size: 103 patients treated continuously with pramipexole for at least one month.
• Who: People with restless legs syndrome at one tertiary hospital in East China, 2016-2018.
• How long: retrospective chart and telephone follow-up.
• Result: 15 of 103 (15%) met Max Planck Institute criteria for augmentation. Those affected had more often switched from another dopaminergic drug (P<0.05), had longer symptom duration before pramipexole (P<0.05), and had longer duration (P<0.05) and higher dosage (P<0.05) of pramipexole.
• Funding: not stated.

Limit of this finding: Retrospective, single centre, 103 people, and augmentation was ascertained partly by telephone interview. The figure is not a controlled incidence and the paper reports p-values without the underlying numbers for most comparisons. We found no Cochrane review or prospective cohort of dopamine agonists and augmentation in restless legs syndrome, so this single retrospective 103-patient series from one centre is the whole of the pramipexole-specific evidence we could source. Fifteen of 103 is 14.6%, which the authors round to 15%; there is no confidence interval and no control group. The comparisons are reported only as P<0.05 thresholds with no effect sizes, so nothing here says how much any factor raises the risk.

Fifteen patients (15%) were classified as having augmentation. Compared to RLS patients without augmentation, more patients with augmentation switched from other dopaminergic drugs (P<0.05) and had a longer duration of RLS symptoms before pramipexole treatment (P<0.05). In addition, patients with augmentation had a longer duration (P<0.05) and higher dosage (P<0.05) of pramipexole than those without augmentation.

The label's own table of dose-related adverse reactions in restless legs syndrome contains rows with no dose gradient and rows where placebo rates were higher than every pramipexole dose. (Source 21)

• Randomized trial, Low certainty.
• Size: 88, 80 and 90 on the three fixed doses versus 86 on placebo.
• Who: People with restless legs syndrome in a 12-week fixed-dose trial.
• How long: 12 weeks.
• Result: Nausea rose with dose (11%, 19%, 27% versus 5% placebo) and so did pain in extremity (3%, 3%, 7% versus 1%) and nasal congestion (0%, 3%, 6% versus 1%). But dyspepsia was 3%, 1%, 4% against 7% on placebo, insomnia 9%, 9%, 13% against 9%, and fatigue 3%, 5%, 7% against 5%.
• Funding: industry (manufacturer's label data)

Limit of this finding: The table is headed 'Dose-Related Adverse Reactions' but several of its own rows show no dose relationship, and for dyspepsia the placebo rate exceeds every pramipexole dose. A reader should not conclude that every reaction listed in this table rises with dose, and should not read the placebo-higher rows as drug effects. The table's stated inclusion rule is also loose: it says it lists reactions occurring in at least 5% of all patients, but dyspepsia never reaches 5% on any pramipexole dose. Table 7 in this label carries no footnote and no footnote markers, so there is nothing further in the source qualifying it.

Table 7 summarizes data for adverse reactions that appeared to be dose related in the 12-week fixed dose study. Table 7 Dose-Related Adverse Reactions in a 12-Week Double-Blind, Placebo-Controlled Fixed Dose Study in Restless Legs Syndrome (Occurring in ≥5% of all Patients in the Treatment Phase) Body System/Adverse Reaction Pramipexole Dihydrochloride Tablets 0.25 mg (N=88) % Pramipexole Dihydrochloride Tablets 0.5 mg (N=80) % Pramipexole Dihydrochloride Tablets 0.75 mg (N=90) % Placebo (N=86) % Gastrointestinal disorders Nausea 11 19 27 5 Diarrhea 3 1 7 0 Dyspepsia 3 1 4 7 Psychiatric disorders Insomnia 9 9 13 9 Abnormal dreams 2 1 8 2 General disorders and administration site conditions Fatigue 3 5 7 5 Musculoskeletal and connective tissue disorders Pain in extremity 3 3 7 1 Infections and infestations Influenza 1 4 7 1 Respiratory, thoracic and mediastinal disorders Nasal congestion 0 3 6 1

What the evidence supports

Starting Parkinson's treatment with pramipexole rather than levodopa left fewer people with dopaminergic motor complications at 23.5 months, but with worse symptom control. (Source 2)

• Randomized trial, Moderate certainty.
• Size: 301 patients (151 pramipexole, 150 carbidopa/levodopa)
• Who: People with early Parkinson's disease who needed dopaminergic therapy, at 22 academic movement disorder sites in the US and Canada.
• How long: 23.5 months.
• Result: Wearing off, dyskinesias or on-off fluctuations in 28% on pramipexole versus 51% on levodopa (hazard ratio 0.45, 95% CI 0.30-0.66; P<.001) - an absolute difference of 23 percentage points, which is about 4 people treated for one fewer person with a motor complication (simple arithmetic on the trial's own percentages; the paper reports no number needed to treat). Mean total UPDRS improvement was greater on levodopa (9.2 versus 4.5 points; P<.001).
• Funding: not stated in the abstract (Parkinson Study Group, multicentre academic)

Limit of this finding: The trial's own figures disagree in one place: this block reports the dopamine-transporter imaging substudy as 39 patients per arm (78 in all), while the paper's methods section describes that substudy as 82 subjects. The published interval is also printed with a stray space, as '0. 30-0.66', and means 0.30 to 0.66. Neither affects the main 28% versus 51% result, but a reader should not treat the imaging substudy numbers as settled.

Initial pramipexole treatment resulted in significantly less development of wearing off, dyskinesias, or on-off motor fluctuations (28%) compared with levodopa (51%) (hazard ratio, 0.45; 95% confidence interval [CI], 0. 30-0.66; P<.001). The mean improvement in 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<.001).

Pramipexole reduced restless legs syndrome symptom scores more than placebo over 12 weeks on both patient-rated and clinician-rated measures. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 344 patients randomised to placebo or one of three fixed doses.
• Who: People with moderate to severe restless legs syndrome.
• How long: 12 weeks (3-week up-titration)
• Result: Adjusted mean change in the International RLS Study Group scale from baseline to week 12: -9.3 (SE 1.0) on placebo versus -12.8, -13.8 and -14.0 on the three pramipexole arms (all p < 0.01). Clinician-rated 'very much' or 'much improved': 51.2% on placebo versus 74.7%, 67.9% and 72.9% - an absolute gain of 17 to 24 percentage points over a placebo response of 51.2%.
• Funding: not stated in the abstract.

Limit of this finding: The three doses did not line up in order of benefit: the proportion rated much or very much improved was 74.7% on the lowest dose (0.25 mg), 67.9% on the middle dose and 72.9% on the highest, so this trial does not show that a higher dose works better. Note also that just over half of the placebo group (51.2%) were rated improved, so most of the improvement people felt on treatment also happened without it.

For IRLS, the adjusted mean (SE) change from baseline to week 12 was -9.3 (1.0) for placebo, -12.8 (1.0) for 0.25 mg/day, -13.8 (1.0) for 0.50 mg/day, and -14.0 (1.0) for 0.75 mg/day (all p < 0.01). Similarly, pramipexole increased the percentage of patients with a CGI-I rating of "very much improved" or "much improved" at the end of the trial (51.2% for placebo and 74.7%, 67.9%, and 72.9% for pramipexole; all p < 0.05)

A single placebo-controlled trial found pramipexole reduced pain and fatigue in fibromyalgia, an off-label use. (Source 6)

• Randomized trial, Low certainty.
• Size: 60 patients randomised 2:1 (pramipexole:placebo)
• Who: People with fibromyalgia at one centre, comorbidities and disability not excluded; about half on narcotic analgesia.
• How long: 14 weeks.
• Result: Visual-analogue pain score fell 36% on pramipexole versus 9% on placebo (treatment difference -1.77 cm). 42% on pramipexole versus 14% on placebo achieved at least a 50% decrease in pain - an absolute difference of 28 percentage points. Total Fibromyalgia Impact Questionnaire treatment difference -9.57. Tender point score, anxiety and depression measures favoured pramipexole but did not reach statistical significance.
• Funding: not stated in the abstract (single-centre investigator trial)

Limit of this finding: One 60-person single-centre trial with 2:1 allocation is not a literature, and concomitant analgesics including narcotics were allowed to continue. The paper gives percentage changes rather than absolute scores for several outcomes. The responder percentages cannot be checked: with 60 people randomised 2 to 1 the arms are 40 and 20, so 42% and 14% are not possible fractions of those groups, and the paper never says how many people were in the analysis. Treat 42% versus 14% as approximate figures from an unreported analysis group.

At 14 weeks, the VAS pain score decreased 36% in the pramipexole arm and 9% in the placebo arm (treatment difference -1.77 cm). Forty-two percent of patients receiving pramipexole and 14% of those receiving placebo achieved > or =50% decrease in pain.

What the evidence does not support

A delayed-start trial found no evidence that pramipexole modifies the course of Parkinson's disease. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 535 patients randomised (261 pramipexole, 274 placebo); 411 analysed at 15 months; 123 in the imaging substudy.
• Who: People diagnosed with Parkinson's disease within 2 years, aged 30-79, at 98 centres.
• How long: 15 months, with placebo recipients switched to pramipexole at 6-9 months.
• Result: Adjusted mean 15-month change in total UPDRS showed no significant difference between early and delayed pramipexole (-0.4 points, 95% CI -2.2 to 1.4, p=0.65). Striatal (123)I-FP-CIT binding fell -15.1% (SE 2.1) with early and -14.6% (2.0) with delayed treatment (difference -0.5 percentage points, 95% CI -5.4 to 4.4, p=0.84).
• Funding: industry-funded (Boehringer Ingelheim GmbH)

Limit of this finding: One set of the paper's own percentages does not match its denominators: 535 people were randomised (261 early, 274 delayed), yet the adverse-event figures are printed as 180 (81%) and 179 (84%), and the serious-event figures as 22 (10%) and 17 (8%). Worked against the randomised numbers those fractions are 69%, 65%, 8.4% and 6.2%. The abstract never says what population the percentages are taken from, so the percentages should not be quoted as rates.

At 15 months (n=411), adjusted mean change in UPDRS total score showed no significant difference between early and delayed pramipexole (-0·4 points, 95% CI -2·2 to 1·4, p=0·65). 62 patients in the early pramipexole group and 61 patients in the delayed pramipexole group were included in the neuroimaging substudy, for which the adjusted mean 15-month change in striatal (123)I-FP-CIT binding was -15·1% (SE 2·1) for early and -14·6% (2·0) for delayed pramipexole (difference -0·5 percentage points, 95% CI -5·4 to 4·4, p=0·84). Overall, 180 (81%) of patients given early pramipexole and 179 (84%) patients given delayed pramipexole reported adverse events (most frequently nausea), and 22 (10%) patients in the early pramipexole group and 17 (8%) in the delayed pramipexole group had serious events, two of which (hallucinations and orthostatic hypotension) were deemed related to study drug. INTERPRETATION: By clinical and neuroimaging measures, pramipexole showed little evidence differentiating 15-month usage from usage delayed for 6-9 months. The results do not support the hypothesis that pramipexole has disease-modifying effects.

In a trial of treatment-resistant bipolar depression, pramipexole did not beat placebo on the 12-week primary outcome and increased hypomanic symptoms. (Source 5)

• Randomized trial, Low certainty.
• Size: 39 participants randomised (18 pramipexole, 21 placebo); 36 in the primary analysis.
• Who: Adults under secondary-care mental health services in England and Scotland with treatment-resistant bipolar depression, on an ongoing mood stabiliser.
• How long: 12 weeks primary outcome, up to 48 weeks follow-up.
• Result: Reduction in self-rated depression score 4.4 (4.8) on pramipexole versus 2.1 (5.1) on placebo: d = -0.72, 95% CI -0.4 to 6.3, p = 0.087. Secondary outcomes at 36 weeks favoured pramipexole (6.28 points, 95% CI 1.85-10.71), as did response (46% versus 6%; p = 0.026) and remission (31% versus 0%; p = 0.030) at trial exit. Hypomania ratings were significantly higher at 12 weeks.
• Funding: independent (NIHR Health Technology Assessment programme, award 16/154/01)

Limit of this finding: The trial closed early and the follow-up period varied, so the 'trial exit' response and remission rates are measured at different times for different people. A reader should not take the significant 36-week and trial-exit results as evidence of benefit when the pre-specified 12-week primary outcome was not met. The 12-week numbers do not hang together as printed: a 4.4-point fall against a 2.1-point fall is a 2.3-point difference, but the confidence interval given is -0.4 to 6.3, which is centred on 2.95. The abstract does not say whether that interval is adjusted and the discrepancy cannot be resolved from it, so the 12-week effect size should be treated as unresolved rather than quoted as a number. One figure differs between the trial's own two publications: the abstract prints the week-36 psychosocial-function confidence interval as 0.38 to 10.35 while the full text's own result prints 0.38 to 10.34. The abstract is the cited source and is recorded as printed; the one-hundredth discrepancy is the publications', not this record's, and changes nothing about the result.

Pramipexole (n = 18) was associated with a greater reduction in QIDS-SR score at 12 weeks versus placebo (n = 21, 4.4 (4.8) vs 2.1 (5.1)): a medium sized (d = -0.72) but not statistically significant difference (95% CI: -0.4 to 6.3, p

Where the evidence is mixed

Across 29 trials, dopamine agonists as a class reduced motor complications compared with levodopa but increased non-motor side effects and controlled symptoms less well. (Source 22)

• Meta-analysis, Moderate certainty.
• Size: 5,247 participants across 29 trials.
• Who: People with early Parkinson's disease randomised to an oral dopamine agonist versus placebo or levodopa.
• How long: varies by trial.
• Result: Dyskinesia odds ratio 0.51 (95% CI 0.43 to 0.59; P<0.00001), dystonia OR 0.64 (0.51 to 0.81), motor fluctuations OR 0.75 (0.63 to 0.90). Against that: oedema OR 3.68 (2.62 to 5.18), somnolence OR 1.49 (1.12 to 2.00), hallucinations OR 1.69 (1.13 to 2.52), and discontinuation for adverse events OR 2.49 (2.08 to 2.98). The review reports odds ratios only, not absolute risks.
• Funding: not stated (Cochrane review)

Limit of this finding: This review pools all oral dopamine agonists, not pramipexole alone, and it says symptom-control data were 'reported too inconsistently and incompletely to meta-analyse' - so the statement that symptom control is poorer with agonists is not itself a pooled estimate. A reader should not read these odds ratios as pramipexole-specific or as absolute risks.

Participants randomised to a dopamine agonist were less likely to develop dyskinesia (odds ratio (OR) 0.51, 95% confidence interval (CI) 0.43 to 0.59; P < 0.00001), dystonia (OR 0.64, 95% CI 0.51 to 0.81; P = 0.0002) and motor fluctuations (OR 0.75, 95% CI 0.63 to 0.90; P = 0.002) than levodopa-treated participants. However, various 'non-motor' side-effects, including oedema (OR 3.68, 95% CI 2.62 to 5.18; P < 0.00001), somnolence (OR 1.49, 95% CI 1.12 to 2.00; P = 0.007), constipation (OR 1.59, 95% CI 1.11 to 2.28; P = 0.01), dizziness (OR 1.45, 95% CI 1.09 to 1.92; P = 0.01), hallucinations (OR 1.69, 95% CI 1.13 to 2.52; P = 0.01) and nausea (OR 1.32, 95% CI 1.05 to 1.66; P = 0.02) were all increased in agonist-treated participants (compared with levodopa-treated participants). Agonist-treated participants were also significantly more likely to discontinue treatment due to adverse events (OR 2.49, 95% CI 2.08 to 2.98; P < 0.00001). Finally symptomatic control of Parkinson's disease was better with levodopa than with agonists, but data were reported too inconsistently and incompletely to meta-analyse.

Where the research disagrees

Whether starting Parkinson's disease treatment with a dopamine agonist rather than levodopa is the better opening move

• Parkinson Study Group (CALM-PD investigators), randomised controlled trial, 301 patients, 23.5 months: Fewer patients receiving initial treatment for PD with pramipexole developed dopaminergic motor complications than with levodopa therapy. Despite supplementation with open-label levodopa in both groups, the levodopa-treated group had a greater improvement in total UPDRS compared with the pramipexole group. (Source 2)
• Stowe and colleagues, Cochrane Review, meta-analysis of 29 trials, 5,247 participants: This meta-analysis confirms that motor complications are reduced with dopamine agonists compared to levodopa, but also establishes that other important side-effects are increased and symptom control is poorer with agonists. (Source 22)

How much

• Reference intake: Dosing is set by the prescriber, not by the reader. As a position, the FDA label (DailyMed SPL for pramipexole dihydrochloride tablets, effective 2026-03-31) sets its own titration schedule for restless legs syndrome - a starting dose taken once daily 2 to 3 hours before bedtime, raised no more often than every 4 to 7 days - and requires the interval between titration steps to be stretched to 14 days in patients with moderate and severe renal impairment, which this section defines as a creatinine clearance of 20 to 60 mL/min. (Source 15)
• Upper limit: As a position, the same label caps the daily amount by indication and by creatinine clearance; the clinical-pharmacology section notes that cimetidine and other drugs cleared by the renal cationic transporter raise pramipexole exposure, which is why the ceiling is lower in kidney impairment. (Source 8)
• Studied: The CALM-PD trial gave pramipexole 0.5 mg three times a day with a levodopa placebo, escalating over the first 10 weeks, versus carbidopa/levodopa 25/100 mg three times a day. (Source 23)
• Studied: The restless legs trial used fixed doses of 0.25, 0.50 and 0.75 mg per day, reached over a 3-week up-titration. (Source 3)
• Studied: PROUD was a delayed-start trial, not a placebo-controlled one: patients were randomised double-blind to pramipexole 1.5 mg a day or to placebo, and the placebo group was assigned to pramipexole at 9 months, or as early as 6 months if considered necessary, with both groups followed for 15 months. The comparison is therefore 15 months of pramipexole against pramipexole delayed by 6 to 9 months. (Source 4)
• Studied: The PAX-BD trial titrated pramipexole over an initial 4 weeks, raising the dose by 0.25 mg a day every 3 days to a target of 2.5 mg a day as tolerated, then held the dose reached through week 12. All of its doses are salt weights. (Source 24)

A common belief, and what the research shows

The belief: Because pramipexole works directly on dopamine receptors, it must protect the dopamine cells that Parkinson's disease destroys.

What the research shows: The delayed-start PROUD trial tested exactly this and found nothing. Starting pramipexole at once rather than 6 to 9 months later made no difference to disability score at 15 months (-0.4 points, 95% CI -2.2 to 1.4, p=0.65) and no difference to striatal dopamine-transporter binding on imaging (-15.1% versus -14.6%, p=0.84). The trial's own words: “The results do not support the hypothesis that pramipexole has disease-modifying effects.” The drug relieves symptoms; nothing shows it protects the cells.

Questions and answers

What is it?

Pramipexole is a man-made drug taken as a tablet that acts directly on dopamine receptors in the brain rather than being turned into dopamine first. It is a non-ergot dopamine agonist, selective for the D2 family of receptors and binding the D3 subtype more tightly than D2 or D4. It comes as immediate-release and extended-release tablets and is used for Parkinson's disease and for moderate-to-severe restless legs syndrome. (Source 1)

What does it do in the body?

It stands in for dopamine at the receptor. In Parkinson's disease the effect is thought to come from stimulating dopamine receptors in the striatum, the part of the brain that has lost its dopamine supply, which improves movement. For restless legs syndrome the label states the precise mechanism is not known, only that the evidence points to the dopamine system being involved. Its D3 preference is often invoked to explain the restless legs effect, but the label says the relevance of D3 binding in Parkinson's disease is unknown. (Source 1)

Is it good or bad for you?

It depends entirely on the setting. In Parkinson's disease, starting with pramipexole instead of levodopa left fewer people with motor complications at 23.5 months (28% versus 51%) but controlled symptoms less well and caused more sleepiness. In restless legs syndrome it reduces symptoms over 12 weeks. Against that, the Cochrane meta-analysis of 5,247 people found dopamine agonists as a class raise oedema, somnolence, hallucinations, constipation and dizziness and more than double the chance of stopping treatment for side effects - so the same drug is a reasonable trade for one person and the wrong drug for another. (Source 22)

How do you get more of it?

Pramipexole is a prescription-only medicine. There is no food, supplement or behaviour that increases it and nothing in the diet contains it. The only way the amount in the body goes up is a prescriber increasing the amount or something slowing its exit through the kidneys: cimetidine raised pramipexole exposure by 50% in a pharmacokinetic study, and drugs sharing the same renal transporter lower its clearance by about 20%. Reduced kidney function does the same. This is information about the drug's handling, not a suggestion to change anything. (Source 8)

If it is harmful, what reduces it?

Pramipexole leaves the body through the kidneys: 90% of a dose is recovered in urine, almost all of it unchanged, and its renal clearance is about three times the rate of filtration, meaning the kidney actively pumps it out. Nothing in the diet speeds that up. If the drug has to come off, the literature is clear that stopping abruptly is the thing to avoid: a withdrawal syndrome of anxiety, panic, depression, sweating, pain and drug craving occurred in 19% of people tapered in one cohort and does not respond to levodopa, so stopping is a prescriber's decision and is done gradually with monitoring. (Source 8)

Why might someone be low in it or missing it?

This question fits a nutrient rather than a drug: nobody is naturally low in pramipexole and there is no deficiency state. The blood level being lower than expected has prosaic causes - doses not taken, or a drug that blocks its effect rather than its level. The label's only drug-interaction entry is that dopamine antagonists - antipsychotics such as phenothiazines, butyrophenones and thioxanthenes, and metoclopramide - may cancel out its effect, so someone can be taking it and get no benefit from it. (Source 7)

Which whole foods contain it or feed it?

No food contains pramipexole; it is a synthetic drug, not a nutrient. Food matters to it in only one way, which is timing: a pharmacokinetic study found a meal does not change how much is absorbed, but delays the peak blood level by about an hour. The label's one food-adjacent warning is about alcohol rather than food - alcohol adds to the drug's sleepiness. (Source 8)

What happens if you do not have it?

Not taking it means not getting its effect, and in restless legs syndrome the size of that effect is measurable: over 12 weeks the International RLS Study Group score fell 9.3 points on placebo and 12.8 to 14.0 points on pramipexole, and 51.2% of people on placebo were rated much or very much improved against 68% to 75% on the drug. So the untreated state is not symptom-free, but a large part of the improvement people experience on treatment also happens on placebo. For Parkinson's disease, going without dopaminergic treatment means the motor disability the drug was started for. (Source 3)

How can you test for it?

There is no routine blood test for pramipexole and no test of whether someone needs it; the decision is clinical. What is monitored is kidney function, because the kidney is what clears the drug and its clearance falls with creatinine clearance - about 75% lower at a creatinine clearance near 20 mL/min and about 60% lower near 40 mL/min than in healthy volunteers, which is why creatinine clearance is used to predict the fall in clearance. Monitoring for the harms - sudden sleep, impulse-control behaviour, hallucinations, augmentation - is by asking, not by assay, and the label stresses that patients may not report these unless directly asked. (Source 8)

References

1. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 12.1 Mechanism of Action. 2026. Read the source
2. JAMA. Pramipexole vs levodopa as initial treatment for Parkinson disease: A randomized controlled trial. Parkinson Study Group.. 2000. PMID 11035889, DOI 10.1001/jama.284.15.1931. Read the source
3. Neurology. Efficacy and safety of pramipexole in restless legs syndrome.. 2006. PMID 16931507, DOI 10.1212/01.wnl.0000231513.23919.a1. Read the source
4. The Lancet Neurology. Pramipexole in patients with early Parkinson's disease (PROUD): a randomised delayed-start trial.. 2013. PMID 23726851, DOI 10.1016/S1474-4422(13)70117-0. Read the source
5. Journal of Psychopharmacology. A randomised double-blind, placebo-controlled trial of pramipexole in addition to mood stabilisers for patients with treatment-resistant bipolar depression (the PAX-BD study).. 2025. PMID 39829389, DOI 10.1177/02698811241309622. Read the source
6. Arthritis & Rheumatism. A randomized, double-blind, placebo-controlled trial of pramipexole, a dopamine agonist, in patients with fibromyalgia receiving concomitant medications.. 2005. PMID 16052595, DOI 10.1002/art.21191. Read the source
7. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 7.1 Dopamine Antagonists. 2026. Read the source
8. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 12.3 Pharmacokinetics. 2026. Read the source
9. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, Patient Information. 2026. Read the source
10. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 5.1 Falling Asleep During Activities of Daily Living and Somnolence. 2026. Read the source
11. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 5.11 Withdrawal Symptoms. 2026. Read the source
12. Archives of Neurology. Dopamine agonist withdrawal syndrome in Parkinson disease.. 2010. PMID 20065130, DOI 10.1001/archneurol.2009.294. Read the source
13. Journal of the Neurological Sciences. Dopamine agonist withdrawal syndrome: A comprehensive review.. 2017. PMID 28104232, DOI 10.1016/j.jns.2016.12.070. Read the source
14. DailyMed (FDA Structured Product Label), Ingenus Pharmaceuticals, LLC. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, EXTENDED RELEASE, section 5.11 Withdrawal Symptoms. 2025. Read the source
15. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 2.3 Dosing for Restless Legs Syndrome. 2026. Read the source
16. Archives of Neurology. Impulse control disorders in Parkinson disease: a cross-sectional study of 3090 patients.. 2010. PMID 20457959, DOI 10.1001/archneurol.2010.65. Read the source
17. Neurology. Falling asleep at the wheel: motor vehicle mishaps in persons taking pramipexole and ropinirole.. 1999. PMID 10371546, DOI 10.1212/wnl.52.9.1908. Read the source
18. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 5.4 Hallucinations and Psychotic-like Behavior. 2026. Read the source
19. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 6.1 Clinical Trials Experience (Restless Legs Syndrome). 2026. Read the source
20. Sleep and Breathing. Augmentation in patients with restless legs syndrome receiving pramipexole therapy: a retrospective study in a single center from China.. 2022. PMID 33864178, DOI 10.1007/s11325-021-02353-9. Read the source
21. DailyMed (FDA Structured Product Label), Torrent Pharmaceuticals Limited. PRAMIPEXOLE DIHYDROCHLORIDE TABLET, section 6.1 Clinical Trials Experience (Table 7, dose-related reactions in RLS). 2026. Read the source
22. Cochrane Database of Systematic Reviews. Dopamine agonist therapy in early Parkinson's disease.. 2008. PMID 18425954, DOI 10.1002/14651858.CD006564.pub2. Read the source
23. JAMA. Pramipexole vs levodopa as initial treatment for Parkinson disease: A randomized controlled trial. Parkinson Study Group.. 2000. PMID 11035889, DOI 10.1001/jama.284.15.1931. Read the source
24. Journal of Psychopharmacology. A randomised double-blind, placebo-controlled trial of pramipexole in addition to mood stabilisers for patients with treatment-resistant bipolar depression (the PAX-BD study). 2025. PMID 39829389, DOI 10.1177/02698811241309622. Read the source