Beruflich Dokumente
Kultur Dokumente
Sleep Medicine
journal homepage: www.elsevier.com/locate/sleep
Original Article
a r t i c l e i n f o a b s t r a c t
Article history: Background: Pramipexole is an effective treatment for restless legs syndrome (RLS), but no controlled
Received 11 June 2010 studies have lasted >12 weeks.
Received in revised form 7 December 2010 Methods: RLS patients (N = 331) with pretreatment serum ferritin >30 ng/mL were randomly assigned to take
Accepted 10 December 2010
double-blind optimized pramipexole (0.125–0.75 mg/d) or placebo for 26 weeks. The primary efficacy
Available online 26 February 2011
endpoint was change in International RLS Study Group Rating Scale (IRLS) score. Other endpoints assessed
global change, symptoms, and QoL. Patients maintained symptom diaries. Cases meeting predefined criteria
Keywords:
for suspected augmentation were reviewed by a blinded expert panel, which used a predefined algorithm.
Restless legs syndrome
Pramipexole
Results: Among 321 patients providing post-baseline data, of whom 234 completed 26 weeks, pramipexole
Dopamine agonists was more effective than placebo by multiple endpoints, including an adjusted mean IRLS score change of
Augmentation 13.7 vs. 11.1 (p = 0.0077) and an IRLS responder rate (P50% score reduction) of 58.6% vs. 42.8%
Therapy (p = 0.0044). Efficacy showed considerable country-to-country variability. Six-Month incidence of confirmed
Long-term treatment augmentation was 9.2% for pramipexole and 6.0% for placebo. The rate increased with treatment duration for
Withdrawal pramipexole but not placebo. Treatment-related adverse events (AEs) were more likely for pramipexole than
for placebo, but discontinuation due to AEs was less likely.
Conclusions: During a 6-month period, pramipexole was effective, safe, and generally well tolerated. Because
risk of augmentation may have increased over 6 months, it should be studied in longer trials. Beginning or mild
augmentation is difficult to distinguish from natural RLS fluctuation, at least in a non-iron-deficient population.
Ó 2011 Elsevier B.V. All rights reserved.
⇑ Corresponding author. Tel.: +43 512 504 23811; fax: +43 512 504 23842. 1. Introduction
E-mail addresses: birgit.ho@i-med.ac.at (B. Högl), dgb@iis.es (D. Garcia-
Borreguero), trenkwalder@pk-mx.de (C. Trenkwalder), ferinistrambi.luigi@hsr.it
Dopaminergic treatment is currently the first-line pharmaco-
(L. Ferini-Strambi), werner.poewe@i-med.ac.at (W. Poewe), stefanie.brenner@
boehringer-ingelheim.com (S.S. Brenner), mandy.fraessdorf@boehringeringelheim. logic intervention in restless legs syndrome (RLS) [1,2]. For moder-
com (M. Fraessdorf), michael.busse@boehringer-ingelheim.com (M. Busse), Stefan. ate to severe cases, the approved agents include pramipexole, a
Albrecht@merz.de (S. Albrecht), richardjhu@mac.com (R.P. Allen). nonergotamine dopamine agonist with high affinity for the D2-like
1
Tel.: +34 91345429; fax: +34 913509593. subfamily of dopamine receptors. In 4 randomized, double-blind
2
Tel.: +49 (561) 6009 200; fax: +49 (561) 6009 126.
3 phase 3 trials [3–6], a total of approximately 1000 patients re-
Tel.: +39 (02) 26433358; fax: +39 (02) 26433394.
4
Tel.: +43 512 504 3850; fax: +43 512 504 3852.
ceived the active drug or placebo for up to 12 weeks. Three of
5
Tel.: +49 7351 54 97894; fax: +49 7351 54 5726. the trials [3–5] had run-in or extension phases lasting up to
6
Tel.: +49 (6132) 77 92914; fax: +49 (6132) 72 92914. 46 weeks, all of which corroborated the benefits documented dur-
7
Tel.: +49 (6132) 77 90693; fax: +49 (6132) 72 90693. ing the double-blind phases. The added phases, however, were
8
Tel.: +49 (0) 69 15 03 1364; fax: +49 (0) 69 15 03 1445. partly or wholly open label, and hence could not definitively assess
9
Tel.: +1 (410) 550 1044; fax: +1 (410) 550 3364.
the treatment’s long-term effects. The trials also were not designed
Deceased.
1389-9457/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.sleep.2010.12.007
352 B. Högl et al. / Sleep Medicine 12 (2011) 351–360
to study augmentation (RLS exacerbation attributed to an ongoing, therefore planned. Testing of mean IRLS change was, by analysis of
initially beneficial RLS treatment) or withdrawal (RLS exacerbation covariance (ANCOVA), adjusted for baseline, country, and treatment
if treatment stops). group. For IRLS, CGI-I, and PGI, responder rates were defined, respec-
Here we report the findings of a phase 4, randomized, double- tively, as the proportion of patients with at least a 50% reduction
blind, placebo-controlled, dose-titration trial with a multiplicity from their baseline total IRLS score, the proportion classified as at
of long-term (26-week) efficacy endpoints. Safety was judged by least ‘‘much improved,’’ and the proportion classifying themselves
profiling adverse events (AEs) and tolerability by rates of prema- as at least ‘‘much better.’’ Among these endpoints, IRLS and CGI-I re-
ture discontinuation. The safety testing also included assessments sponder rates were the trial’s key secondary outcomes. All respon-
for incidence of augmentation, by submission of all suspected cases der-rate findings were subjected to Cochran–Mantel–Haenszel
to a blinded expert panel, and for incidence of withdrawal, by eval- test, and all other efficacy findings to van Elteren test (stratified by
uation of RLS severity after each patient’s end of treatment. country). In all analyses, p < 0.05 was considered statistically significant.
Adverse events were identified by asking each patient at each
2. Methods trial visit, ‘‘How have you felt since your last visit?’’ Mild AEs were
defined as being easily tolerated, moderate AEs as interfering with
At 42 sites in 9 European countries, adults (aged 18–85 years) usual activities, and severe AEs as preventing usual activities. Seri-
with idiopathic RLS were randomly assigned, at a 1:1 ratio, to receive ous AEs were defined as being life-threatening or resulting in death
double-blind pramipexole or placebo. For entry, all patients were re- or significant incapacity, requiring hospitalization, or requiring
quired to meet all diagnostic criteria of the International RLS Study medical or surgical intervention to prevent such outcomes. All
Group (IRLSSG) [7], to have a baseline total score >15 on the Study AEs occurring within 48 h after a patient’s last intake of trial med-
Group’s International RLS Rating Scale (IRLS) [8], and to have expe- ication were classified as having happened under treatment. After
rienced RLS symptoms at least 2–3 days per week throughout the weeks 4, 12, and 26, supine and standing pulse rate and diastolic
prior 3 months. Patients were excluded for serum ferritin 630 ng/ and systolic blood pressure were measured.
mL, known hypersensitivity to pramipexole, augmentation during Because augmentation would be identifiable only in reference
previous RLS treatment, unsuccessful previous treatment with non- to initial response to treatment, all patients were asked to maintain
ergotamine dopamine agonists (e.g., pramipexole, ropinirole), any an RLS symptom diary for the week preceding randomization and
non-RLS sleep disorder, any major Diagnostic and Statistical Manual the week preceding each scheduled evaluation. For each hour of
of Mental Disorders, Fourth Edition, Text Revision psychiatric disorder each of those weeks, the diary documented the severity of a pa-
within the prior 2 years, change in any antidepressant regimen tient’s RLS symptoms (none; mild/nonbothersome; bothersome),
within the prior 4 weeks (or any anticipated change), and any use, their time of onset, and the patient’s status (sleeping, awake and
within the prior 2 weeks, of dopamine agonists, levodopa, or any active, awake but at rest). At all scheduled evaluations, investiga-
medication or dietary supplement capable of altering RLS tors administered an augmentation questionnaire enabling them
symptoms. Women with childbearing potential were excluded for to screen for suspected augmentation by four standards: IRLSSG
pregnancy, inadequate contraception, or current breastfeeding of a criteria [7], developed at a National Institutes of Health workshop
child. Patients with daytime RLS symptoms were not excluded. and published in 2003; current international criteria (Max-
The study was approved by the independent ethics committees of Planck-Institute criteria) [12], developed at a European RLS Study
the participating countries. Informed, written consent was obtained Group consensus conference and published in 2007; a score P5
from all patients prior to randomization. on the 3-item Augmentation Severity Rating Scale (ASRS) [13];
During the first 4 weeks, treatment was optimized. For pramip- and an investigator’s own judgment, including reliance on diary
exole recipients, treatment commenced with a week at 0.125 mg/ data. A positive finding on any of these was cause for case referral
d. After weeks 1, 2, 3, and 4, stepwise dosage adjustment was per- to an independent, 4-member expert panel. In accordance with
mitted to 0.125, 0.25, 0.50, or 0.75 mg/d (pramipexole dihydro- current criteria [12], only patients treated for P4 weeks could be
chloride monohydrate), based in each patient on clinically considered. To ensure that case referrals were based on thorough,
sufficient response (on the Patient Global Impression [PGI] scale uniform knowledge of augmentation, all investigators had under-
[9]) and tolerability. Treatment then continued at the patient’s gone training by the expert panel’s chairperson.
optimized dosage until the end of week 26. All patients were asked For each referred case, the expert panel, blinded to the subject’s
to take their treatment once daily, 2–3 h before expected bedtime. treatment group, reviewed all available data using a predefined
After weeks 1, 4, 6, 12, 18, and 26, all patients underwent on- algorithm (Table 1), so as to achieve a consensus in confirming or
site evaluation of treatment efficacy, treatment compliance, AEs, excluding augmentation or in judging the data to be insufficient
and augmentation. At each of these visits, patients assessed them- for a decision. In the absence of consensus, the panel’s chairperson
selves by IRLS (the trial’s primary outcome measure) and PGI, and made the final decision. The counts referred to below as ‘‘classified
were assessed on the Clinical Global Impressions-Global Improve- augmentation’’ include not only confirmed augmentation but also
ment (CGI-I) scale [9]. For each IRLS assessment, each patient was all referred cases judged to have insufficient data. The counts re-
asked to read each of the instruments 10 items and provide the ferred to as ‘‘confirmed augmentation’’ include only cases in which
required self-rating on a patient worksheet while the examiner re- the data were judged to be sufficient.
mained available to clarify any misunderstandings a patient may Withdrawal was defined as a posttreatment increase (i.e., wors-
have had and to record the patient’s responses on an investigator ening) of P4 points on a patient’s IRLS score, as measured
worksheet. Self-assessment tools also included the Johns Hopkins 7 ± 1 days after cessation of treatment. The comparison was with
Restless Legs Syndrome Quality of Life Questionnaire (RLS-QoL) the higher of the patient’s scores at screening and randomization.
[10] and the RLS-6 set of scales [11]. All investigators received
training on study procedures including all scales and worksheets
and the patient diary (described below). 3. Results
The trial’s sample size was based on detection of treatment effi-
cacy. For 26-week change in mean IRLS total score, documentation 3.1. Study subjects
of a 4.5-point difference between pramipexole and placebo at a 2-
sided 5% significance level with 90% power would require that each Of 331 randomized patients, 329 received treatment (starting,
group have 151 patients. Recruitment of 160 per group was for the first subjects, in May 2007, and ending, for the last subjects,
B. Högl et al. / Sleep Medicine 12 (2011) 351–360 353
Table 1
The expert panel’s algorithm for confirming or excluding augmentation.
1. Review investigator’s basis for suspicion. Which standard(s) did the patient meet, and was the suspicion plausible?
2. Review initial treatment efficacy. By current augmentation criteria [12], there must initially have been benefit
3. Review relation to dosage. Did worsening emerge during dosage increase? (However, by current augmentation criteria [12],
only worsening occurring in patients treated P4 weeks could be considered.)
4. Review IRLS data. Was there a change consistent with augmentation?
5. Review diary data. Were there changes consistent with augmentation?
6. Review QoL data. Was QoL consistent with augmentation?
Table 2
in July 2008). In their baseline characteristics (Table 2, top), the
Patients’ baseline characteristics.
166 pramipexole recipients resembled the 163 placebo recipients,
including a mean IRLS score in the range (21–30) signifying severe Pramipexole Placebo
RLS. Within these groups, 162 and 159 patients contributed post- Number of patients treated 166 163
baseline IRLS/PGI/CGI-I/RLS-6 data, and 157 and 153 contributed Age, y; mean (SD) 57.9 (12.7) 55.8 (11.4)
Females, % 61.4 57.7
postbaseline RLS-QoL data. On all scales, the pramipexole group
RLS duration, y; median (P25%, P75%) 1.45 (0.0, 8.2) 1.20 (0.0, 7.2)
resembled the placebo group at baseline (Table 2, bottom). Prema- RLS duration, y; mean (SD) 6.0 (9.6) 5.4 (8.5)
ture discontinuation was less frequent for pramipexole than for Naïve to RLS treatment, % 80.1 76.1
placebo, at 21.1% (35/166) vs. 36.8% (60/163). Trial completion IRLS total score, mean (SD) 23.9 (5.3) 23.5 (5.4)
rates were therefore 78.9% (131/166) for pramipexole and 63.2% Number of patients providing postbaseline 157–162 153–159
(103/163) for placebo. data
RLS-6
Sleep dissatisfaction,a median (P25%, P75%) 6.0 (4.0, 8.0) 6.0 (4.0, 8.0)
3.2. Study-drug exposure Severity while falling asleep, median (P25%, 5.0 (3.0, 7.0) 5.0 (2.0, 7.0)
P75%)
Nighttime severity, median (P25%, P75%) 5.0 (2.0, 7.0) 5.0 (2.0, 7.0)
Among pramipexole recipients, 27 (16.3%) were maintained on Daytime severity while resting, median 4.0 (2.0, 6.0) 4.0 (2.0, 6.0)
0.125 mg/d, 53 (31.9%) on 0.25 mg/d, 44 (26.5%) on 0.5 mg/d, and (P25%, P75%)
42 (25.3%) on 0.75 mg/d. In the placebo group, maintenance ‘‘dos- Daytime severity while active, median 1.0 (0.0, 2.0) 1.0 (0.0, 2.0)
age’’ tended to be higher, with 17 patients (10.4%) on ‘‘0.125 mg/d,’’ (P25%, P75%)
Daytime tiredness/sleepiness, median (P25%, 4.0 (2.0, 6.0) 4.0 (2.0, 6.0)
29 (17.8%) on ‘‘0.25 mg/d,’’ 57 (35.0%) on ‘‘0.5 mg/d,’’ and 60
P75%)
(36.8%) on ‘‘0.75 mg/d.’’ RLS-QoL, median (P25%, P75%) 72.5 (60.0, 70.0 (57.5,
82.5) 82.5)
3.3. IRLS/CGI-I/PGI IRLS, International Study Group RLS Rating Scale; QoL, quality of life; RLS, restless
legs syndrome; SD, standard deviation.
a
Termed dissatisfaction here, so as to emphasize that on all RLS-6 scales, high
Among pramipexole recipients, IRLS total score (SE) decreased
scores represent severe detriment.
by an adjusted mean 13.7 (0.8), vs. 11.1 (0.8) for placebo
(p = 0.0077). The decrease was significant after week 1 and re-
mained so throughout the trial (Fig. 1), with a maximum difference
from placebo of 3.7 points at week 6. Country by country, however,
the 26-week mean score change showed a treatment-group differ- Weeks After Randomization
ence (Table 3) ranging from 7.4 points favoring pramipexole in 1 4 6 12 18 26
0
Germany to 1.5 points favoring placebo in the Netherlands. In con-
Adjusted Mean Change (SE)
Table 3
26-Week IRLS score change, by country.
CI, confidence interval; IRLS, International RLS Rating Scale; SD, standard deviation.
Table 4
recipients to experience AEs and were also more likely to experi-
CGI-I responder ratea at 26 weeks, by country.
ence treatment-related AEs. Rates of premature discontinuation
Pooled country Pramipexole Placebo due to AEs, however, were lower for pramipexole than for placebo.
n Responder rate (%) n Responder rate (%) AE frequencies are summarized by medical type in Table 6. Among
Austria + Slowakia 8 75.0 7 57.1 the three types with a frequency >10% in either treatment group,
Belgium + Spain 10 80.0 10 60.0 nausea was almost four times more common and fatigue was
Finland 26 69.2 26 50.0 slightly more common for pramipexole than for placebo, while
Germany 42 73.8 41 36.6
headache was less common for pramipexole than for placebo.
Ireland 11 63.6 13 46.2
Netherlands 23 73.9 21 66.7 The study’s 11 serious AEs were hospitalizations (of 4.8% of pram-
United Kingdom 42 57.1 41 53.4 ipexole and 1.8% of placebo recipients), of which only 1 (orthostatic
hypotension and vomiting in a placebo recipient) was deemed to
CGI-I, Clinical Global Impressions-Global Improvement scale.
a
Proportion of group classified as at least ‘‘much improved.’’
be treatment-related. Overall, vital signs (including supine and
standing systolic and diastolic blood pressure) showed no notable
changes from baseline in either treatment group.
Table 5
Frequencies of adverse events: summary.
3.7. Augmentation
Category, n (% of group) Pramipexole Placebo
group group
Among 152 pramipexole and 149 placebo recipients treated for
Patients treated 166 (100.0) 163 (100.0)
P4 weeks, and therefore eligible for augmentation diagnosis, 47
Proportion with any AE 120 (72.3) 106 (65.0)
Proportion with treatment-related AEs 64 (38.6) 50 (30.7) (30.9%) and 40 (26.8%) were referred for expert-panel assessment
Proportion with severea AEs 17 (10.2) 15 (9.2) of suspected augmentation. Of these referrals, 18 (11.8% of eligible
Proportion with seriousb AEs 8 (4.8) 3 (1.8) pramipexole recipients) and 14 (9.4% of eligible placebo recipients)
Proportion with AEs leading to 19 (11.4) 23 (14.1)
were classified as augmentation cases, encompassing either con-
withdrawal
firmed augmentation or insufficient data for a definitive decision
AEs, adverse events. (Fig. 2). Among the 32 classified cases, mean (SD) exposure to
a
Defined as blocking the patient from work or other usual activities. study drug was 170.6 (29.8) days for pramipexole, compared with
b
Defined as fatal, life-threatening, significantly disabling, or requiring hospi-
talization.
127.2 (70.3) for placebo.
Randomized:
Table 6 331
Adverse events reported by P5% of either treatment groupa. Randomized but not treated:
2
MedDRA preferred term, n (% of group) Pramipexole Placebo group Treated:
group 329
5 before the study’s final visit, only 5 stopped treatment—2 (of 10)
on pramipexole and 3 (of 8) on placebo.
4
New Confirmed Cases
3.8. Withdrawal
only RLS symptom frequency and intensity but also RLS impact on Together, the available data suggest some guidance for evalua-
daily life, mood, tiredness, and sleep. tion of suspected augmentation. In general, for dopamine agonists
The present trial was designed to investigate withdrawal as a such as pramipexole, augmentation should not be expected within
safety parameter. In general, withdrawal is considered most likely the first 3 months of treatment, at least in patients with a serum
to develop after high dosage. Here, however, the frequency of post- ferritin level >30 ng/mL and no history of augmentation on dopa-
treatment IRLS score deterioration by P4 points was independent mine agonists. Hence, events reported during the first 3 months
of dosage. The deterioration could therefore be interpreted as ver- should be carefully assessed as possible random RLS symptom fluc-
ifying loss of benefit if an effective treatment stops [5]. Neverthe- tuation. With increasing duration of treatment, augmentation rates
less, the treatment-group difference (of 10.5% among can be expected to increase (while RLS fluctuation may not in-
pramipexole recipients vs. 1.5% among placebo recipients) sug- crease). Indeed, within pramipexole’s approved dosage range, and
gests that pharmacodynamic mechanisms may be involved. for serum ferritin >30 ng/mL, treatment duration appears to be
The present trial was also designed to investigate augmentation more important than treatment dosage as an augmentation risk
as a facet of safety, utilizing placebo control (making this the first factor, at least during the first 6 months.
pramipexole trial to have done so). Augmentation is perhaps the Lastly, cases of augmentation developing during the first
most clinically relevant complication of long-term dopaminergic 6 months appear likely to be mild, at least in patients with serum
treatment of RLS [18]. In the present trial, its 6-month incidence ferritin >30 ng/mL. Mildness would make early cases all the harder
on pramipexole resembled the rate for placebo, both for classified to differentiate from ‘‘noise,’’ e.g., RLS fluctuation, or, in clinical tri-
augmentation, a highly conservative analysis in which all cases als, the varying diagnostic accuracy in different countries or at dif-
involving inadequate information were counted as augmentation ferent clinical centers. This does not preclude the possibility that
(11.8% vs. 9.4%), and for confirmed augmentation (9.2% vs. 6.0%). augmentation severity as well as prevalence increases with dura-
This resemblance might reflect an RLS sample too small to calcu- tion of treatment, even at dosage not exceeding an approved max-
late augmentation rates, especially given the length of the study. imum. For researchers, the present trial’s most salient implication
As noted above, the sample size was based on statistical testing may be that a fuller understanding of the clinical impact of aug-
for efficacy, not augmentation (which was not a primary endpoint, mentation will have to be based on comparative RLS-intervention
and for which there was no previous experience to guide in design- trials lasting substantially longer than 6 months.
ing controlled trials).
It should be added that in a clinical trial, assessment of augmen- Disclosures
tation may be hampered by fixed maximum dosage (at which level
a patient’s fulfillment of augmentation criteria might signify only Birgit Högl has been a consultant for Boehringer Ingelheim (BI),
insufficient RLS treatment). It may likewise be hampered by the GlaxoSmithKline (GSK), and UCB (for RLS); has served on advisory
artificiality of an expert panel not only blinded to treatment (for boards for BI, UCB, Cephalon, Merz, Jazz, Pfizer, and Lundbeck; and
the sake of placebo control) but also unable to interview patients. received speaker honoraria from BI, UCB, Cephalon, GSK, Pfizer,
In the present trial, the recorded rates might also reflect the use of and Novartis. Diego Garcia-Borreguero received honoraria for advi-
a serum ferritin level 630 ng/mL as an exclusion criterion. The sory boards or lectures from Boehringer Ingelheim, GlaxoSmithK-
present trial is alone among published trials in employing this va- line, UCB Pharma, Pfizer, Xenoport, Sanofi-Aventis, Jazz, and MSD.
lue. In a large 30-week study of levodopa vs. cabergoline for RLS Stefanie Brenner, Michael Busse, and Mandy Fraessdorf are employ-
[19], lower ferritin at baseline was a risk factor for occurrence of ees of Boehringer Ingelheim. Stefan Albrecht was an employee of
augmentation during the trial, and in a recently published series Boehringer Inghelheim at the time of this study and approved the fi-
of 302 consecutive RLS patients [20], those with current augmen- nal draft of this paper. Claudia Trenkwalder is a consultant for Nov-
tation had lower ferritin values (although the mean was within artis and Mundipharma. She is on the advisory board of UCB and
the normal range) than those without augmentation. Boehringer Ingelheim (BI), and has received honoraria from UCB,
Even so, the present trial’s 6-month rates of augmentation BI, GlaxoSmithKline, Pfizer, and Teva. She has also provided expert
resembled those in previous, retrospective studies of pramipexole testimony for Mundipharma and Axxonis. Richard P. Allen has
for RLS, among which Ferini-Strambi [21] reported a rate of 8.3% in received research support from GlaxoSmithKline, NIH, and Pharma-
patients treated at least 6 months and Silber et al. [22] a rate of cosmos, and consultation fees, honoraria, or travel support from
approximately 9% at 6 months. For its part, the similarity between GlaxoSmithKline, Boehringer Ingelheim, Jazz, UCB, Xenoport, Luit-
rates of augmentation for pramipexole and for placebo suggests pold, Pharmacosmos, Pfizer, EMD-Serono, Neurogen, Novartis, and
that beginning or mild augmentation can be difficult to distinguish Orion Pharma. Luigi Ferini-Strambi received honoraria from serving
from natural symptom fluctuation occurring in untreated RLS. on the scientific advisory board of UCB Pharma, Boehringer Ingel-
The critical question is whether or not the likelihood of random heim, GlaxoSmithKline, Sanofi-Aventis, and Transept Pharmaceuti-
fluctuation, which can be mistaken for augmentation, remains sta- cals. Werner Poewe has served on the advisory board of and has
ble over time, while rates of augmentation on medication increase received consultancy and lecture fees (including honoraria) from
with duration of treatment. The present study’s augmentation- Eisai, Teva, Novartis, GlaxoSmithKline, Boehringer Ingelheim, UCB,
incidence findings show this general pattern. Moreover, a relation- Orion Pharma, Merck Serono, and Schering Plough in relation to clin-
ship between augmentation and duration of treatment is ical drug development programs for Parkinson’s disease.
consistent with prior studies of open-label RLS therapy. For
open-label pramipexole, Silber et al. [22] showed (in a Kaplan– Conflicts of Interest
Meier plot) that augmentation occurred in about 6–8% of patients
over the first 6 months of treatment, and in another 12–14% over The ICMJE Uniform Disclosure Form for Potential Conflicts of
the next 6 months. The study reported overall rates of 20% for Interest associated with this article can be viewed by clicking on
1 year and 33% for a mean 27.2 months. Similarly, Winkelman the following link: doi:10.1016/j.sleep.2010.12.007.
et al. [23] reported a rate of 32% for a mean 21.2 months. In a re-
cent 6-month study of levodopa for RLS [24], augmentation also
occurred progressively, with a rate of 60% at 6 months (and no Acknowledgement
indication that the rate would not have continued to increase with
duration of treatment). This study was funded by Boehringer Ingelheim GmbH.
B. Högl et al. / Sleep Medicine 12 (2011) 351–360 357
Appendix (continued)
Total number of centres: 42
Centre Investigator name and Affiliation and address Number of
number role patients enrolled
per centre
4902 Bergtholdt – PI Emovis GmbH, Bereich Klinische 34
Filler – Investigator Forschung, Dr. med. Bettina
Novy – Investigator Bergtholdt, Wilmersdorfer
Straße 79, 10629, Berlin
4903 Ehret – PI Dr. med. Reinhard Ehret, 4
Friedrich – Investigator Schloßstr. 29
2163 Berlin-Steglitz
4904 Gussmann – PI ClinPharm Intern. GmbH 17
Effenberg – Investigator Studienzentrum, Berlin
Stössel – Investigator Dr. med. Manfred Gussmann
Janusz-Korczak-Str. 8
12627 Berlin (Hellersdorf)
4905 Stahl – PI ClinPharm International GmbH 15
Degtyareva – Investigator Dr. med. Hans-Detlev Stahl
Sigal – Investigator Johannisplatz 1, 04103 Leipzig
Streck – Investigator
4906 Simonow - PI Alexander Simonow 2
Neurologische Praxis
Hauptstrasse 106
35745 Herborn
4907 Klein – PI Medizinisches Studienzentrum 5
Helbig – Investigator Dr. med. Martin Klein
Oehler – Investigator Augustinerstr. 15
97070 Würzburg
4908 Schumann – PI Neurologie, Psychiatrie, 9
Bitter – Investigator Psychotherapie, Dr. med. Günther
Schumann, Castroper Hellweg
422, 44805 Bochum
4909 Benes – PI Somni bene, Institut für 20
Meissner – Investigator Medizinische Forschung und
Schlafmedizin Schwerin GmbH
Dr. med. Heike Benes
Arsenalstr. 10, 19053 Schwerin
Appendix (continued)
Total number of centres: 42
Centre Investigator name and Affiliation and address Number of
number role patients enrolled
per centre
3403 Poza – PI Hospital de Donostia, Neurology 6
Department, Paseo Dr.
Beguiristain, s/n
20014 San Sebastián
3405 Paniagua – PI Hospital Virgen de las Nieves 2
Sanchez – Investigator Neurophysiology Department
Pabellón de Traumatología
Carretera de Jaén, s/n
18014 Granada
3406 Ordoño – PI Hospital Arnau de Vilanova 6
Chornet – Investigator Neurophysiology Department
San Clemente, 12
46015 Valencia
References manual for psychopharmacology, revised ed. Rockville, MD: NIMH USA; 1976.
p. 216–22.
[10] Abetz L, Vallow SM, Kirsch J, Allen RP, Washburn T, Earley CJ. Validation of the
[1] Silber MH, Ehrenberg BL, Allen RP, et al. An algorithm for the management of
restless legs syndrome quality of life questionnaire. Value Health 2005;8:
restless legs syndrome. Mayo Clin Proc 2004;79:916–22.
157–67.
[2] Oertel WH, Trenkwalder C, Zucconi M, et al. State of the art in restless legs
[11] Kohnen R, Oertel WH, Stiasny-Kolster K, Benes H, Trenkwalder C. Severity
syndrome therapy: practice recommendations for treating restless legs
rating of restless legs syndrome: validation of the RLS-6 scales. Sleep
syndrome. Mov Disord 2007;22:S466–75.
2004;27(Suppl.):A304. Abstract 680.
[3] Partinen M, Hirvonen K, Jama L, et al. Efficacy and safety of pramipexole in
[12] Garcia-Borreguero D, Allen RP, Kohnen R, et al. Diagnostic standards for
idiopathic restless legs syndrome: a polysomnographic dose-finding study—
dopaminergic augmentation of restless legs syndrome: report from a World
the PRELUDE study. Sleep Med 2006;7:407–17.
Association of Sleep Medicine-International Restless Legs Syndrome Study Group
[4] Oertel WH, Stiasny-Kolster K, Bergtholdt B, et al. Efficacy of pramipexole in
consensus conference at the Max Planck Institute. Sleep Med 2007;8:520–30.
restless legs syndrome: a six-week, multicenter, randomized, double-blind
[13] Garcia-Borreguero D, Kohnen R, Högl B, et al. Validation of the augmentation
study (effect-RLS study). Mov Disord 2006;21:343–53.
severity rating scale (ASRS): a multicentric, prospective study with levodopa
[5] Trenkwalder C, Stiasny-Kolster K, Kupsch A, Oertel WH, Koester J, Reess J.
on restless legs syndrome. Sleep Med 2007;8:455–63.
Controlled withdrawal of pramipexole after 6 months of open-label
[14] Baker WL, White CM, Coleman CI. Effect of nonergot dopamine agonists on
treatment in patients with restless legs syndrome. Mov Disord 2006;21:
symptoms of restless legs syndrome. Ann Fam Med 2008;6:253–62.
1404–10.
[15] Fulda S, Wetter TC. Where dopamine meets opioids: a meta-analysis of the
[6] Winkelman JW, Sethi KD, Kushida CA, et al. Efficacy and safety of pramipexole
placebo effect in restless legs syndrome treatment studies. Brain 2008;131:
in restless legs syndrome. Neurology 2006;67:1034–9.
902–17.
[7] Allen RP, Picchietti D, Hening WA, et al. Restless legs syndrome: diagnostic
[16] de la Fuente-Fernandez R, Lidstone S, Stoessl AJ. Placebo effect and dopamine
criteria, special considerations, and epidemiology. A report from the restless
release. J Neural Trans Suppl 2006;70:415–8.
legs syndrome diagnosis and epidemiology workshop at the National Institutes
[17] Perlis ML, McCall WV, Jungquist CR, Pigeon WR, Matteson SE. Placebo effects in
of Health. Sleep Med 2003;4:101–19.
primary insomnia. Sleep Med Rev 2005;9:381–9.
[8] Walters AS, LeBrocq C, Dhar A, et al. Validation of the International Restless
[18] Allen RP, Earley CJ. Augmentation of the restless legs syndrome with
Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med
carbidopa/levodopa. Sleep 1996;19:205–13.
2003;4:121–32.
[19] Trenkwalder C, Högl B, Benes H, Kohnen R. Augmentation in restless legs
[9] National Institute of Mental Health (NIMH). Early clinical drug evaluation unit
syndrome is associated with low ferritin. Sleep Med 2008;9:572–4.
(ECDEU). Clinical global impressions. In: Guy W, editor. ECDEU assessment
360 B. Högl et al. / Sleep Medicine 12 (2011) 351–360
[20] Frauscher B, Gschliesser V, Brandauer E, et al. The severity range of restless [23] Winkelman JW, Johnston L. Augmentation and tolerance with long-term
legs syndrome (RLS) and augmentation in a prospective patient cohort: pramipexole treatment of restless legs syndrome (RLS). Sleep Med 2004;5:
association with ferritin levels. Sleep Med 2009;10:611–5. 9–14.
[21] Ferini-Strambi L. Restless legs syndrome augmentation and pramipexole [24] Högl B, Garcia-Borreguero D, Kohnen R, et al. Progressive development of
treatment. Sleep Med 2002;3(Suppl.):S23–5. augmentation during long-term treatment with levodopa in restless legs
[22] Silber MH, Girish M, Izurieta R. Pramipexole in the management of restless syndrome: results of a prospective multi-center study. J Neurol 2010;257:
legs syndrome: an extended study. Sleep 2003;26:819–21. 230–7.