Abstract
Background: Open clinical trials indicate that low doses of pergolide, a long-acting D1 and D2 dopamine agonist, lead to a reduction in the symptoms of restless legs syndrome (RLS) with subjective improvement in sleep quality.
Objective: To assess the therapeutic efficacy of pergolide in improving sleep and subjective measures of well-being in patients with idiopathic RLS using polysomnography and clinical ratings.
Methods: In a randomized, double-blind, placebo-controlled crossover design we enrolled 30 patients with idiopathic RLS according to the criteria of the International RLS Study Group. All patients were free of psychoactive drugs for at least 2 weeks before the study. Patients were monitored using polysomnography, clinical ratings, and sleep diaries at baseline and at the end of a 4-week pergolide or placebo treatment period. The initial dosage of 0.05 mg pergolide was increased to the best subjective improvement paralleled by 20 mg domperidone tid.
Results: At a mean dosage of 0.51 mg pergolide as a single daily dose 2 hours before bedtime, there were fewer periodic leg movements per hour of time in bed (5.7 versus 54.9, p < 0.0001), and total sleep time was significantly longer (373 versus 261 minutes, p < 0.0001). Ratings of subjective sleep quality, quality of life, and severity of RLS were improved significantly without relevant adverse events.
Conclusion: Pergolide given as a single low-to-medium bedtime dose in combination with domperidone provides a well-tolerated and effective treatment of sensorimotor symptoms and sleep disturbances in patients with primary RLS.
Clinical features of restless legs syndrome (RLS) include disagreeable sensations, experienced mostly in the lower limbs and rarely in the arms, and motor restlessness caused by an irresistible urge to move. Typically, symptoms occur primarily in the evening or at night when at rest, with temporary relief by activity. In 1995, the International RLS Study Group established diagnostic criteria, which include the major clinical symptoms as the four minimal criteria.1 Polysomnographic studies revealed that more than 80% of affected patients have involuntary periodic leg movements (PLMs),2 which may occur during sleep or relaxed wakefulness, or both.3,4 It has been suggested that an abnormal modulation of sensory information through the diencephalo-spinal dopaminergic system may play a role in the pathogenesis of RLS.5 In idiopathic RLS, only symptomatic therapy is possible. Based on the proven therapeutic efficacy of levodopa6,7 and bromocriptine8 we hypothesized that pergolide, a potent and long-acting dopamine D1- and D2-receptor agonist, might be of therapeutic value in idiopathic RLS. Recent open clinical trials indicate that pergolide is effective in RLS, ameliorating sensory symptoms and improving subjective sleep quality.9-11 The present study was designed as a placebo-controlled, double-blind trial to assess the therapeutic efficacy of pergolide in improving sensorimotor symptoms, total sleep time, and subjective measures of quality of sleep and well-being in patients with idiopathic RLS using polysomnography and several clinical ratings. These hypotheses were defined a priori in the study protocol.
Methods.
Patient selection.
The participants in the study were recruited from the outpatient clinic for movement and sleep disorders at the Max Planck Institute of Psychiatry in Munich and the Department of Neurology at the University of Marburg, Germany. The protocol was approved by two independent ethics committees at the Max Planck Institute and the University of Marburg. Written informed consent to participate in the study was obtained from all patients before they entered the study. Patients agreed not to take any psychotropic drugs (e.g., levodopa, dopamine agonists, benzodiazepines, opioids, carbamazepine, antidepressants, and neuroleptics) at least 2 weeks before baseline and during the entire trial (except study medication). Before entering the study, all patients underwent a complete neurologic and general medical examination, including serum chemistry laboratory tests, EKG, and chest radiograph. In addition, polysomnography, including the monitoring of oronasal airflow, respiration effort, and oxygen saturation, was performed on two successive nights to rule out other specific sleep disorders such as sleep-related breathing disorders.
Inclusion criteria.
Patients were included if they fulfilled the diagnostic criteria of the International RLS Study Group1: 1) an irresistible urge to move associated with sensory complaints of the lower limbs, 2) motor restlessness, 3) worsening of the symptoms at rest with at least partial and temporary relief by activity, and 4) increased severity in the evening or at night. In addition, the symptoms had to have been stable for 2 weeks before the study. Baseline polysomnography had to reveal an abnormal PLM index (more than five PLMs per hour of time in bed). Based on clinical experience in the treatment of RLS, a sleep onset latency longer than 25 minutes or a sleep efficiency of less than 85%, or both, were also defined as baseline polysomnography inclusion criteria. Serum chemistry (complete blood count, creatinine, electrolytes, liver enzymes, bilirubin, urea, uric acid, lactate dehydrogenase, glucose, vitamin B12, folic acid, iron, ferritin, basal thyroid stimulating hormone, T3, and T4) and urine analysis had to be in the normal range.
Exclusion criteria.
Patients younger than 18 and older than 70 years and patients with signs of secondary RLS or any other specific primary sleep disorder such as sleep-related breathing disorder (apnea/hypopnea index >5 per hour during total sleep time, baseline polysomnography) were excluded from the study. Subjects with a history of alcohol or drug abuse were also excluded as were pregnant or lactating women and women without safe contraception. Finally, patients were excluded for any of the following reasons: established or suspected hypersensitivity to ergot derivatives; treatment with an antihistamine at the time of evaluation; clinically significant or unstable medical conditions including serious cardiovascular, pulmonary, or renal disease; active gastric or duodenal ulcer or a history of ulcers in the previous 12 months; concurrent or past diagnosis of malignant melanoma.
Design and randomization.
The study was designed as a 12-week, double-blind, randomized, placebo-controlled clinical trial with two crossover periods. The study protocol is given in figure 1. After informed consent was received, patients were assigned a study number in consecutive order, and they entered a 1-week baseline period for assessment of inclusion/exclusion criteria. In the course of the study, seven visits to the outpatient clinic and phone calls were required to note subjective measures of efficacy and to assess any adverse events of the study drug.
Figure 1. Study protocol. Visit 1: Informed consent, medical history, physical examination, EKG, chest radiograph, clinical laboratory variables, inclusion/exclusion criteria, study admission, dispensing of study medication. Visit 2: Polysomnography (PSG), Clinical Global Impressions scale (CGI), visual analogue scales (VAS), sleep questionnaire, rating of severity of restless legs syndrome (RLS), sleep diary, adverse events, study-drug record. Visit 3: Randomization, rating of RLS severity, adverse events, dispensing of study medication, study-drug record. Visits 3t and 5t (telephone interviews): Control of up-titration, rating of RLS severity, adverse events. Visits 4 and 6: PSG, CGI, VAS, sleep questionnaire, rating of RLS severity, sleep diary, adverse events, dispensing of study medication, study-drug record. Visit 5: Rating of RLS severity, sleep diary, adverse events, dispensing of study medication, study-drug record. Visit 7: Physical examination, clinical laboratory variables, rating of RLS severity, adverse events, study-drug record.
Dosing.
The study capsules contained either pergolide (0.05 mg) (Eli Lilly and Company, Indianapolis, IN) or placebo (0.25 mg). Unused capsules were returned to the project coordinator, who kept a record of all medication dispensed to and returned by the patients. Patients took pergolide or placebo once a day, 2 hours before bedtime. During the first 2 weeks of both treatment periods, patients used a titration scheme starting with 0.05 mg and increasing the dosage by 0.05 mg per night (increase by 0.10 mg only at night 8) until the best improvement in the patient’s opinion was achieved. This dosage remained constant for the subsequent 2-week treatment period. The maximum dosage was 0.75 mg. To prevent peripheral dopaminergic side effects, all patients took 20 mg domperidone tid unblinded throughout the entire study beginning with the baseline period.
Polysomnography.
Polysomnographic recordings of each patient were performed at baseline, with one night of adaptation, and at the end of both treatment periods. An Electroencephalograph Digital 24 polygraph (Schwarzer, Munich, Germany) was used by the center in Munich, and the center in Marburg employed an Alice 3 polygraph (Healthdyne Technologies Inc., Marietta, GA). Patients were not allowed to nap during the day before the sleep investigation. Sleep was recorded between 11 pm and 7 am, including EEG (C3-A2, C4-A1, C3-C4, using the International 10-20 System12), an electro-oculogram, a submental EMG in accordance with standardized guidelines,13 and an EMG of both anterior tibialis muscles. Sleep recordings were scored visually by trained raters using the Rechtschaffen and Kales criteria13 in both centers. Sleep onset latency was defined as the time from lights off to the first epoch of stage 2 sleep lasting longer than 3 minutes, and REM latency was defined as the time between sleep onset and the first epoch of REM sleep. Computation of the other sleep measures has been described in detail elsewhere.14 PLMs were scored during both sleep and wakefulness in accordance with international scoring rules.15 The total number of PLMs, the number of PLMs per hour of time in bed (PLM index), and the number of PLMs during sleep (PLMS) associated with arousals per hour of total sleep time (PLMS arousal index) were calculated. Arousals included K-complexes, alpha activity, or theta waves accompanied by an EEG frequency shift of 3 seconds or longer in accordance with the guidelines of the American Sleep Disorders Association.16 Arousals or leg movements after single respiratory events (apneas or hypopneas) were excluded from the calculations.
Subjective ratings.
Sleep questionnaire.
Patients completed a validated sleep questionnaire (SF-A, Goertelmeyer17), which consists of 22 items relating to the sleep process, on the evening before and the morning after polysomnography. The items were combined into five scales. The quality-of-sleep scale (9 items) was used as a primary efficacy criterion.
Sleep diaries.
To measure the effects of treatment on sleep quality, patients were asked to fill in a sleep diary during the last week of each treatment period (weeks 5 and 11) and the day before and after polysomnography.
Quality of life questionnaire.
Quality of life was measured during the last week of each treatment period using visual analogue scales (modified 50-mm Hamburger Visual Analogue Scales18), which can be subdivided into two categories—“life satisfaction” and “negative feelings and complaints.”
Severity ratings.
Patients rated the severity of RLS at baseline and at the end of each treatment period using an 11-point scale. They rated the severity of RLS while falling asleep, during the night, and on the following day.
Physicians’ rating of RLS.
Disease severity, changes in severity during the study, therapeutic results, and adverse events were rated by a physician at the end of each crossover period with the Clinical Global Impressions scale19 (CGI).
Safety evaluation.
The safety of pergolide therapy was assessed by examination of the frequency and type of adverse events, clinically relevant changes in laboratory data, and premature discontinuation of study participation. Adverse event recording was based on spontaneous reporting and made use of clear-cut definitions of their classifications (serious versus nonserious), intensity, relationship to drug treatment, frequency, course, and need for therapeutic intervention.
Efficacy variables.
The primary efficacy variables were the PLM index during sleep and wakefulness, total sleep time, and subjective sleep quality. The secondary outcome measures included other polysomnographic variables (number of PLMs, PLMS arousal index, sleep latency, number of awakenings, percentage of sleep stages), CGI ratings, and additional ratings made by the patients (quality of life, sleep diary). Calculation of the sample size was based on the results of a controlled crossover study7 that investigated the efficacy of levodopa/benserazide in patients with RLS using comparable methods.
Statistical analysis.
For the primary statistical analysis, a per-protocol analysis was performed; eligible patients must have completed the initial assessment and the 5- and 11-week assessments. It was the aim of the confirmatory statistical analysis to show the superiority of pergolide over placebo using the three primary efficacy criteria. All variables were baseline-adjusted, i.e., the differences between the measures in crossover periods 1 and 2 and the initial values at baseline were calculated. To control for the experimental error introduced by multiple testing of three hypotheses, the Z-transformed measures of the three primary end points were combined into a total efficacy indicator, as proposed by O’Brien.20 The null hypothesis of no differences between the pergolide and placebo groups in the global efficacy measure was tested approximately against the alternative of pergolide being superior to placebo at α = 0.05 using crossover tests as proposed by Lehmacher.21 To test for carryover effects, the sum of periods 1 and 2 was determined for the two sequences and the values compared using the Mann-Whitney U test. Period effects, i.e., differences between the first and second crossover period, were evaluated by comparing the two crossover periods. Carryover and period effects were not found in the primary efficacy analyses. Because the null hypothesis was rejected for the global efficacy measure, all pairs of two primary efficacy criteria, and then all efficacy measures individually, were evaluated within a closed test procedure proposed by Lehmacher et al.,22 with an experimentwise type I error of α = 0.05. Descriptive statistics for pergolide and placebo efficacy measures were reported as pooled data from both crossover periods. The p values, however, represent the results from the crossover analysis described earlier. In addition, the number of PLMs during sleep and wakefulness were calculated separately for 1-hour periods of the night and compared by Student’s t-test in the crossover analysis. The statistical analysis was done with the Statistical Analyses Systems, version 6.11 (SAS Institute, Cary, NC).
Results.
Subjects.
Twenty-eight patients (12 men, 16 women; mean age 57.2 ± 8.9 years, range 28 to 70) completed the study. Severity of RLS was rated by an experienced physician as “markedly ill” (on average) using the severity item of the CGI.19 Ten of the 28 patients had never been treated for restless legs symptoms. The clinical characteristics of the subjects are given in table 1.
Clinical characteristics of patients who completed the study*
Of the 36 Caucasian patients initially screened for the study, 6 were excluded before randomization: 3 patients withdrew their consent, 2 were excluded because of sleep apnea syndrome, and 1 was excluded because of severe bradycardia. Of the 30 randomized and treated patients, 2 dropped out prematurely: a 28-year-old withdrew consent 7 days after the start of crossover period 1 (treatment with pergolide) and a 58-year-old woman dropped out because of epigastric pain 7 days after the start of crossover period 1 (treatment with placebo). The trial profile is given in figure 2.
Figure 2. Trial profile.
Dosage.
The mean pergolide dosage at the end of each crossover period was 0.51 ± 0.18 mg (range 0.25 to 0.75), and the mean placebo dosage was 0.69 ± 0.15 mg (range 0.25 to 1.10) (p = 0.003). The dosage was 0.5 mg or less in 17 patients treated with pergolide but in only 1 patient during placebo treatment. Eight patients on pergolide took a dosage of 0.75 mg, the maximum specified in the protocol, as did 17 patients on placebo. Two patients on placebo took more (0.10 and 0.35 mg per night) in the second crossover period than the maximum specified.
Efficacy.
Crossover analysis of the three primary end points revealed that pergolide was superior to placebo in the reduction of the PLM index, prolongation of total sleep time, and improvement of subjective sleep quality. For details see tables 2 and 3⇓.
Treatment effects: Polysomnographic sleep measures (mean ± standard deviation)
Treatment effects: Subjective and Clinical Global Impression (CGI) ratings (mean ± standard deviation)
Polysomnography.
Sleep measures.
Table 2 gives detailed information on how the two treatments affected sleep continuity and sleep architecture. Mean sleep efficiency improved from 55 to 78% with pergolide (p = 0.0001). Sleep onset latency was reduced from 45 to 25 minutes; however, this difference did not reach the level of significance. The amount of wake time was less with pergolide (p = 0.0001), but the mean total number of nocturnal awakenings (spontaneous and PLM-associated) was similar in the two treatment groups (pergolide 23.5, placebo 22.0). The number of PLMS associated with full awakenings was significantly less with pergolide (2.3 versus placebo 8.9; p < 0.0001). Regarding sleep architecture, pergolide treatment led to an increase in stage 2 sleep (p < 0.0001) but to a decrease in slow wave sleep (SWS) (p = 0.025). The amount of REM sleep did not differ between the groups.
Periodic leg movements.
The mean absolute number of PLMs during sleep and wakefulness was much smaller during pergolide treatment than it was during placebo (45.7 versus placebo 438.0; p < 0.0001). Furthermore, both the PLM index (p < 0.0001) and the PLMS arousal index (p < 0.0001) were significantly lower (see table 2). As shown in figure 3, PLMs were reduced throughout the night. The PLM index related to wakefulness and different sleep stages is shown in table 2.
Figure 3. Time course of the mean frequency of periodic leg movements (PLM) during sleep and wakefulness calculated separately for 1-hour periods. Values are mean ± SEM.
Patients’ ratings.
Sleep questionnaire (SF-A).
Quality of sleep was rated significantly better with pergolide (see table 3).
Sleep diaries.
Patients reported fewer nocturnal awakenings, a shortened sleep onset latency, and an increase in the estimated total sleep time for the last week of each treatment period. Before and after polysomnography, patients reported a significant reduction in their urge to move with pergolide when falling asleep, during the night, and during the day (see table 3).
Severity ratings.
On the RLS severity scale, patients on pergolide had mean severity scores below 1 for symptoms of restless legs during the period of falling asleep, during the night, and during the day. Their ratings were significantly higher when they were on placebo (p < 0.0001) (see table 3).
Quality of life questionnaire.
Pergolide showed better results in both the categories “life satisfaction” and “negative feelings and complaints” (see table 3).
Physicians’ ratings of severity of RLS.
In the efficacy-related scales of the CGI, pergolide was rated as superior to placebo. The therapeutic effect was rated better for pergolide but was nearly unchanged for the placebo condition (see table 3).
Safety.
Adverse events.
Under treatment with pergolide, the most frequent adverse events were nausea (n = 12), headache (n = 6), and rhinitis (n = 6). Vomiting occurred in five patients, abdominal pain and dizziness in four each, and abnormal vision, diarrhea, and rash in three each. Two patients reported dry mouth and one patient vivid dreams. Nineteen of these events were seen as probably related to therapy. Five events (vomiting in three patients and nausea and dizziness in one each) were rated as severe. However, no adverse event was so severe that the affected patient chose to drop out of the study. During placebo treatment, the most frequent adverse events were headache (n = 9), abdominal pain (n = 7), constipation (n = 4), nausea (n = 3), and rash (n = 3). Two adverse events (abdominal pain, dyspnea) were rated as severe. One of them (abdominal pain) was severe enough that the patient stopped participating in the study.
Physicians’ global assessment of tolerability.
On the CGI, tolerability was rated worse for patients receiving pergolide than it was for those receiving placebo, reflected in the higher frequency of adverse events with pergolide (see table 3).
Discussion.
The results of this double-blind, randomized, controlled study show that pergolide is highly effective in treating sensorimotor symptoms and sleep disturbances in RLS. The polysomnographic data revealed that a 4-week treatment regimen with a single mean dosage of 0.51 mg pergolide daily 2 hours before bedtime significantly reduced the number of PLMs, led to normal values on both the PLM index and the PLMS arousal index, and prolonged total sleep time for approximately 2 hours. In parallel with the polysomnographic data, patients reported an improvement in quality of life and subjective sleep quality with a decrease in the severity of restless legs symptoms during the night and the following day. The consistent improvements in all major clinical features of RLS provide compelling evidence and support the validity of the results. However, because of the short treatment period, we cannot provide any information about the efficacy in long-term treatment with pergolide. The results of our study confirm previous reports9-11,23 that low-to-medium doses of pergolide are effective in the treatment of RLS. Another placebo-controlled study, recently published in abstract form, also showed that pergolide is effective in reducing PLMs as measured by actigraphy.24
Although total sleep time was markedly improved, there was a reduction in the amount of SWS. These findings indicate that relatively low doses of pergolide may alter the distribution of non-REM sleep stages toward more light sleep. The higher number of spontaneous arousals during pergolide treatment may be a result of this postulated effect. Pharmacologic studies in humans and animals have shown that the dopaminergic system is involved in the modulation of sleep and waking. In rats the dopamine-receptor agonists apomorphine, bromocriptine, and pergolide induced biphasic effects depending on the doses administered.25 Small doses of apomorphine and bromocriptine decreased wakefulness and increased SWS and REM sleep, whereas higher doses induced the opposite effects. Pergolide also showed biphasic effects, but it suppressed REM sleep at both low and high doses.25 The effects of dopaminergic agents on sleep are complex and seem to depend on the species, the doses, and the receptor specificity of the drug.26-28 However, a decrease in wakefulness after small doses of dopamine-receptor agonists is common. It has been postulated that this effect is related to an inhibition of dopaminergic pathways probably via the activation of presynaptic inhibitory D2 receptors.29
Pergolide is a semisynthetic ergoline dopamine agonist with potent activity at presynaptic D2 and postsynaptic D1 and D2 dopamine receptors.30 The time to mean peak plasma concentration of pergolide ranges from 2.4 to 2.7 hours.31 Because the peak of PLMs in RLS patients is assumed to be between midnight and 1 am,32 drug intake between 9 pm and 10 pm should ameliorate symptoms of RLS most effectively. By measuring the time course of PLM activity we could demonstrate that the efficacy of a single dose of pergolide administered daily at 9 pm was maintained throughout the night, probably due to the long elimination half-life (mean 17 hours) of the drug. Patients even reported that the positive effects were still present the next day.
The mean daily maintenance dose of pergolide in our study (0.51 mg) was higher than that reported by Silber et al.10 (mean 0.26 mg) or Earley and Allen9 (mean 0.37 mg), which may be explained by a difference in the severity of the restless legs symptoms. In addition, domperidone coadministered to prevent peripheral side effects allowed the use of higher and more effective pergolide dosages. However, because no validated severity scale is yet available, it is difficult to compare the results from different studies. Most adverse events resulting from pergolide were mild and tolerable, and no specific interventions were necessary.
Dopaminergic drugs are the treatment of choice in RLS. Controlled studies have demonstrated that levodopa in combination with a peripheral decarboxylase inhibitor is effective in treating patients with idiopathic and uremic RLS.7,33 However, an end-of-dose morning rebound phenomenon34 and an earlier appearance of RLS symptoms in the evening, or even in the afternoon, with levodopa (which has been termed “augmentation”34) may cause major problems. Allen and Earley35 found that this phenomenon was severe enough to require medication change for 50% of patients, but was less frequent with pergolide treatment.9 This could be confirmed by a recent study that demonstrated the benefits of pergolide in RLS patients who had developed augmentation with levodopa treatment.11 An earlier appearance of RLS symptoms did not occur in patients in the present study, but the duration of treatment was only 4 weeks. Long-term studies with different drugs may elucidate the question of whether this phenomenon is a drug-related alteration in receptor-mediated mechanisms or an alteration in the patients’ circadian rhythm of restless legs symptoms.
Acknowledgments
Acknowledgment
The authors thank Elisabeth Kappelmann, RN, for her excellent technical assistance.
Footnotes
-
See also pages 907, 932, 938, 1060, and 1064
-
Eli Lilly and Company, USA, provided the drugs for this study. Lilly Deutschland GmbH, Germany, provided a $25,000 grant for a part-time research nurse. Neither Eli Lilly and Company, USA, nor Lilly Deutschland GmbH paid any pertinent financial support to the authors or their immediate families.
-
Presented in part at the 50th annual meeting of the American Academy of Neurology; Minneapolis, MN; April 28, 1998.
- Received August 3, 1998.
- Accepted November 7, 1998.
References
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.