Abstract
Objectives: To assess the therapeutic efficacy, required dose, and tolerability of pregabalin in patients with idiopathic restless legs syndrome (RLS).
Methods: This was a double-blind, placebo-controlled trial with polysomnographic control, providing Class II evidence. Ninety-eight patients underwent a 2-week single-blind period with placebo; 58 were randomized to receive pregabalin or placebo for 12 weeks under a flexible-dose schedule. Endpoints were mean change from baseline in the International Restless Legs Scale (IRLS) total score, Clinical Global Impression (CGI), and RLS-6 scales, as well as changes in periodic limb movements (PLMs) and sleep architecture.
Results: Patients under treatment with pregabalin had a greater improvement in IRLS score than under placebo (63% vs 38.2%; p < 0.05). The mean effective dose of pregabalin at the end of treatment was 322.50 mg/day (±98.77), although therapeutic effects were already seen at a mean dose of 139 mg/day. Similarly, improvements were observed on the CGI, RLS-6 scale, and the Medical Outcomes Study sleep scale (all p < 0.01) when compared to placebo. Treatment with pregabalin also resulted in a reduction of the mean (±SD) PLM index (p < 0.001). Furthermore, there was a marked improvement in sleep architecture with an increase in slow wave sleep (p < 0.01), and decreases in wake after sleep onset and stages 1 and 2 (p < 0.05). Pregabalin was generally well-tolerated. Adverse events were mild but common, and included unsteadiness, daytime sleepiness, and headache.
Conclusions: This study shows significant therapeutic effects of pregabalin on both sensorial and motor symptoms in restless legs syndrome. Treatment with pregabalin was associated with an improvement of sleep architecture and periodic limb movements. Adverse events included unsteadiness and sleepiness and should be screened carefully in the working population, particularly when pregabalin is administered in the afternoon.
Classification of evidence: This study provides Class II evidence that pregabalin is effective for the treatment of restless legs syndrome and improves sleep architecture and periodic limb movements in placebo-unresponsive patients.
Glossary
- AE=
- adverse event;
- ANCOVA=
- analysis of covariance;
- BL=
- baseline;
- CGI=
- Clinical Global Impression;
- CGI-S=
- Clinical Global Impression Severity;
- CI=
- confidence interval;
- DA=
- dopaminergic agent;
- ECG=
- electrocardiogram;
- IMP=
- investigational medicinal product;
- IQR=
- interquartile range;
- IRLS=
- International Restless Legs Scale;
- MOS=
- Medical Outcomes Study;
- NREM=
- non-REM sleep;
- PLM=
- periodic limb movement;
- PLMI=
- periodic limb movement index;
- PSG=
- polysomnography;
- RLS=
- restless legs syndrome;
- RR=
- relative risk;
- SE=
- standard error;
- STAI=
- State Trait Anxiety Inventory;
- SWS=
- slow wave sleep;
- WASO=
- wake time after sleep onset.
In recent years, a number of drugs have been studied for the treatment of restless legs syndrome (RLS); of these, only dopaminergic agents (DAs) have received regulatory approval in both the United States and Europe. However, this class of agents has potential limitations that should be taken into account during their routine clinical use:
-
Not all patients with RLS respond to DAs.1,2
-
Not all patients tolerate DAs.3,4
-
DAs might lose therapeutic efficacy over time.5
-
Treatment of RLS with DAs may often lead to augmentation, which is a specific increase in RLS severity during long-term treatment. Although its incidence is not well-established, substances like pramipexole might show rates of up to 33% after 30 months of treatment.2
-
Neuropathic forms of RLS might respond better to anticonvulsants than to DAs.6,7
-
During treatment with DAs, sleep architecture still shows a reduction of slow wave sleep (SWS) and an increase in wake time after sleep onset (WASO) compared to the healthy age-matched population.8
There is, therefore, a clinical need for drugs that could be easily used in primary care.9 Gabapentin and its prodrug, gabapentin enacarbil, have been shown to be effective for the treatment of RLS,6,10,11 and therefore, α-(∝)2-δ ligands of the voltage-dependent calcium channel could play a potential role in treating RLS.6,12,13
Pregabalin, another drug of the same class, is effective for the treatment of neuropathic pain,14 and also provides a consolidation of sleep in these patients.15 Furthermore, in an open-label case series, pregabalin has been shown to have a therapeutic effect in mainly neuropathic RLS,16 but studies in patients with idiopathic RLS are lacking. It was the aim of the present study to examine under double-blind conditions the therapeutic effectiveness, effective dosage at which maximal therapeutic effects are obtained, therapeutic effects on periodic limb movements (PLMs) and sleep architecture, and safety and tolerability of pregabalin in idiopathic RLS.
METHODS
Design.
This was a randomized, multicenter, double-blind, placebo-controlled, parallel-group, flexible-dose study, providing Class II evidence, that sought to examine the efficacy, safety, and tolerability of pregabalin for the treatment of idiopathic RLS over a 12-week period. The study was performed between 2006 and 2008 in 2 large sleep clinics in Spain with extensive experience in RLS.
Subjects.
Men and women aged 18–80 years with idiopathic RLS (International Restless Legs Scale [IRLS] total score17 ≥15 points at baseline) that interfered with sleep onset or sleep maintenance on ≥4 nights/week for at least 6 months were included in the study. Diagnosis was made through a thorough examination of medical history, followed by a physical examination. Blood count and blood chemistry (including serum levels for creatinine, iron, ferritin, and transferrin) were performed along with a determination of urinary creatinine clearance.
Exclusion criteria were any form of secondary RLS, coexistence of severe medical or psychiatric disorders, previous treatment lasting >12 weeks with DAs, serum ferritin <10 μg/L, severe comorbid sleep disorders that might confound assessment, or shift work. Before the screening visit, patients underwent a 2-week washout period for any medications with CNS effects.
Protocol.
Standard protocol approvals, registrations, and patient consents.
The study was approved by both local institutional review boards; written informed consent was obtained from all patients participating in the study.
Placebo run-in.
Following the screening visit, a single-blind placebo run-in was performed. Patients who had a >40% improvement in their IRLS total score during this period were considered placebo responders and discontinued from the study.
Baseline visit.
At the baseline visit, severity of RLS was assessed using the IRLS rating scale, Clinical Global Impression (CGI),18 and the RLS-6 scale.19 In addition, patients were evaluated by means of the Medical Outcomes Study (MOS) sleep scale20 and the State Trait Anxiety Inventory (STAI).21 At the baseline visit, patients were investigated for inclusion and exclusion criteria and were required to have an IRLS total score >15 points. Furthermore, in order to be able to assess the efficacy of pregabalin on PLMs, patients with a PLM index (PLMI) <10/hour of recording time were excluded from the study.
Treatment schedule.
Patients were randomized 1:1 to receive either pregabalin or placebo at a dose of 150 mg daily at 9:00 pm. Study dose adjustments were performed weekly and were based on clinical judgment of their efficacy and tolerability. Flexible-dose titration was performed according to the following steps: baseline visit: 150 mg (1 capsule at 9:00 pm); week 1: 300 mg (1 capsule at 9:00 pm); week 2: 300 mg or 450 mg (2 capsules ×300 mg, 1 × 150 mg at 9:00 pm); week 3: 300 mg or 450 mg (1 capsule at 9:00 pm; if daytime symptoms were present and the patient was already receiving 450 mg, then 1 75-mg capsule was added at 2:00 pm); week 4: 300 mg or 450 mg (2 capsules [300 mg and 150 mg] at 9:00 pm, 1 capsule 75 mg or 150 mg at 2:00 pm); week 6: same as week 4; week 8: final dose adjustment if necessary. No further adjustments were allowed between weeks 8 and 12 (end of treatment). No other medication with known effects on RLS was allowed during the study.
The study medication was discontinued after a 4-day downtitration schedule. A follow-up visit was performed 2 weeks after the final visit.
Outcome measures.
Efficacy assessments.
The primary outcome measure in both groups was change in the IRLS score between baseline and week 12.
The secondary outcome measures were change between baseline and week 12 in CGI Severity (CGI-S) scale18; RLS-6 scale19; and MOS scale.
The MOS scale includes questions on subjective perception of sleep initiation, sleep maintenance, sleep quality, daytime somnolence, and sleep breathing disorders.22
Polysomnography.
PSG studies were performed between 11:00 pm and 7:00 am at baseline and final visit (week 12). Sleep stages were scored according to standard American Academy of Sleep Medicine criteria.23
Tolerability assessments.
Adverse events (AEs) were rated at baseline and at every subsequent visit. A blood test (blood count and biochemistry including iron levels), electrocardiogram (ECG), and urinalysis were carried out at baseline and at the end of the treatment period.
Randomization and allocation.
Before trial initiation, the biostatistician executed a previously validated statistical program to obtain the randomization list using blocks of 4 numbers to ensure equal size among comparison groups. The final randomization list was not accessed until the clinical database was locked.
At the baseline visit, once eligibility was confirmed, subjects were randomized 1:1 to 1 of the 2 blinded treatment groups. Blinded treatment numbers were sent in a list to the investigator sites. The subject identification and treatment randomization numbers remained the same throughout the study and were not reallocated to other subjects. The bulk medication was received from Pfizer and consisted of pregabalin capsules (75, 150, and 300 mg) and the equivalent in placebo. The investigational medicinal product (IMP) labeling and packaging followed a highly defined process, in accordance with currently applicable laws and regulations, and ensured that the blinded nature of the trial was respected. The bottles of pregabalin or placebo with the 3 study doses were the same size. After these procedures, the IMP was stored in a local authorized depot and distributed to the trial site pharmacies.
Statistical analyses.
The t test or median test were used to compare continuous variables (expressed as means ± SD or median and interquartile range [IQR]) while categorical variables were expressed as percentages. Proportions were compared by the χ2 test or the Fisher test. The null hypothesis was rejected in each statistical test when p < 0.05. Analysis was performed using SPSS.v.15.0 software.
Change in IRLS total score from baseline (visit BL) to week 12 was analyzed using an analysis of covariance (ANCOVA) with the change score as the dependent variable and the treatment condition and baseline IRLS score as independent variables. Assumptions for the ANCOVA model were checked using plots of predicted values vs residuals as well as plots of the baseline score vs change score for each treatment.
Justification of sample size.
Sample size for this study assumed the results of a previous crossover study6 which had shown an 8-point difference between gabapentin and placebo. The SD of 8.6 points was based on a conversion of the SD reported and was based on standard errors (SE), Ns, and an assumed intrapatient correlation of 0.5 for IRLS scores, s2 = se2/(1 − R), with Ns = 25 and SE = 1.3. With 28 subjects per group, the study had at least 90% power to detect a clinically significant difference between treatment arms. The test assumed a type I error of 0.05 with 2-sided testing. Table e-1 on the Neurology® Web site at www.neurology.org shows the estimated effects of treatment on primary and secondary outcome indices.
RESULTS
Patients.
Ninety-eight subjects diagnosed with idiopathic RLS were enrolled in the study. From this group, 58 (24 men, 34 women) met the aforementioned inclusion criteria and were randomized to either the active treatment group (n = 30) or the placebo group (n = 28) (figure 1). Baseline demographics and RLS history were similar between the 2 groups (table 1).
Figure 1 Study flow chart
AE = adverse event; ITT = intent-to-treat; OSAS = obstructive sleep apnea syndrome; PLMI = periodic limb movement index.
Table 1 Baseline demographics and restless legs syndrome history
Single-blind, placebo run-in, screening period.
Forty patients were excluded during the screening period. Of these, 8 were excluded because of placebo response. During this same run-in phase, the IRLS score improved in the 58 patients who were finally randomized. However, placebo response during the placebo run-in occurred evenly among groups: for the group that was later randomized to pregabalin, the IRLS score improved from a mean score (SD) of 22.07 (3.69) to 19.8 (4.16), vs 23.46 (3.58) to 21.46 (3.81) for the group that remained under placebo (figure 2).
Figure 2 Total IRLS scores from screening to week 12 (p = 0.022)
BL = baseline; IRLS = International Restless Legs Scale.
Rating scales.
International RLS Study Group rating scale.
Figure 2 shows the total score of the IRLS rating scale over the study period. The mean IRLS scores at baseline were 19.80 for the pregabalin and 21.46 for the placebo group. The IRLS mean total score was lower following the 12-week treatment period with pregabalin than with placebo (mean ± SD: 6.85 ± 6.87 vs 11.2 ± 8.60; p < 0.005). Furthermore, the baseline adjusted absolute difference between week 12 and baseline was 4.92 points larger under pregabalin than under placebo (95% confidence interval [CI] 0.73–9.12). In the pregabalin group, there was also a greater relative reduction in the total score by 26.54% (95% CI 6.49–46.58) compared to placebo (both p = 0.01). During the first week of treatment (at a fixed dose of 150 mg/day), patients in the pregabalin group reached a score of 12.2 (±6.69) in contrast to 17.46 (±5.21) for those in the placebo group (p < 0.001). Following discontinuation of treatment, IRLS scores increased in the pregabalin group from 6.85 (±6.86) to 15.6 (±7.16) (p < 0.01), while the placebo group did not show a significant change 11.21 (±8.60) vs 10.36 (±7.94).
Remission of RLS symptoms (an IRLS total score <10 at week 1224) was achieved in 73.3% of patients treated with pregabalin vs 39.3% of those treated with placebo (p = 0.009). A total of 73.3% of pregabalin-treated patients were responders; that is, their total IRLS score decreased by at least 50% by week 12 vs 39.3% for the placebo group (relative remission of 86.7%; 95% CI 25.5–147.8; p = 0.016 or RR [relative risk] 1.87; 95% CI 1.12–3.10). Nine (30%) patients achieved full remission (an IRLS score of 0 points at the end of treatment) in the pregabalin group compared with 4 (14.3%) in the placebo group (p < 0.01).
Stratification of the patient's sample according to baseline severity on the IRLS scale did not show differences in the improvement of IRLS score. Neither were differences seen when the degree of improvement of those patients describing pain among their symptoms at baseline (n = 5 in each treatment group) was compared to those who did not.
Clinical Global Impression of change.
Following 12 weeks of treatment, 88.9% of patients in the pregabalin group reported “any improvement,” and 11.1% showed “no changes.” Corresponding percentages in the placebo group were 63.6% and 36.4% (χ2: 4.44, p < 0.035).
RLS-6 scale.
As shown in table e-1, compared to placebo, pregabalin had an important effect on sleep (item 1) (p < 0.001), RLS when falling asleep (item 2) (p < 0.01), RLS at night (item 3) (p < 0.05), as well as marginal differences on daytime RLS at rest (item 4) (p < 0.1).
Medical Outcomes Study scale.
Compared to placebo, pregabalin improved MOS scores for sleep disturbance (p < 0.001), sleep adequacy (p = 0.001), and sleep quantity (p < 0.001).
State Trait Anxiety Inventory.
No significant differences were observed between the 2 treatment groups on the STAI scores.
Sleep studies.
Periodic limb movements.
Compared to placebo, pregabalin significantly improved all 3 measurements of motor dysfunction: PLMI (p = 0.001), PLM arousal index (p < 0.05), and PLM during wakefulness index (p < 0.05) (table 2).
Table 2 Sleep and mean periodic limb movement indices
Sleep indices.
As can be seen in table 2, pregabalin significantly improved non-REM sleep (NREM) (p = 0.02), percentage of stage 1 sleep (p = 0.021), percentage of stage 2 sleep (p = 0.025), percentage of stage 3 sleep (p = 0.031), minutes of stage 4 sleep (p = 0.001), and WASO (p < 0.01). Furthermore, SWS (stages 3 and 4) significantly improved in the pregabalin group, while there was a worsening in the placebo group (p < 0.001). No changes were seen in total sleep time, sleep efficiency, sleep latency, REM sleep, or in the apnea-hypopnea index.
Dose.
The mean daily dosage of pregabalin at the end of treatment was 337.50 (±105.6) mg/day, while the mean final dose of placebo was 409.82 (±105) mg/day (p < 0.01). The mean effective dose to improve PLMI was 300 mg/day.
Safety and reasons for discontinuation.
For safety and reasons for discontinuation, see table 3. Unsteadiness and daytime sleepiness were common under pregabalin (p < 0.05). Their severity was mild at 300 mg but more severe in subjects taking doses over 450 mg/day (which included daytime administration). Daytime sedation occurred mainly in subjects taking doses over 450 mg/day. Other common AEs, although not significantly different between groups, were headache, blurred vision, mouth dryness, vertigo, dyslalia, and body fluid retention. These AEs were in general mild: within the pregabalin group, 4 patients discontinued the study due to AEs, and 2 due to lack of efficacy. Within the placebo arm, 8 patients discontinued due to lack of efficacy; 1 was lost to follow-up. The rate of AEs under pregabalin increased with the dose: 9% of all AEs occurred at 150 mg, 71.7% at 300 mg, and 12.1% at 450 mg/day. No cases of augmentation were seen.
Table 3 Adverse events in both treatment arms
DISCUSSION
Our data show that pregabalin is effective over 12 weeks for the treatment of RLS symptoms, as measured on specific rating scales. The rates of response and remission following treatment were greater with pregabalin compared with placebo. In addition to improving the neurosensitive symptoms, pregabalin improved motor dysfunction during the night, as shown by all PLM indices. Pregabalin also increased NREM sleep time, with a reduction in stage 1, an increase in stage 2 sleep and notably, SWS. Such changes in objective evaluations were also confirmed by major improvements in the MOS scale. The general view is that in RLS a high PLMI can, either alone or in association with persistent neurosensitive symptoms, lead to sleep disruption. However, the degree to which the improvement in motor dysfunction leads by itself to a normalization of sleep is still a matter of controversy. Hence, a normalization of the PLMI during sleep under dopaminergic agents generally does not result in an improvement of sleep parameters such as SWS time or arousal index.8,25,26 Taken together the above data suggest that pregabalin produces changes during sleep beyond mere improvement of motor dysfunction and causes a clear improvement in sleep architecture.
The mean effective dose at which these therapeutic effects were obtained was 337 mg/day. However, given that the study design included a flexible-dose approach, it is likely that lower doses are sufficient to reach a therapeutic effect. On the other hand, 23.3% of patients required a dose of 450 mg/day.
Pregabalin was more effective than placebo despite a high placebo response. The responder rate, arbitrarily defined as an improvement of >50% in the IRLS scale total score, was present in up to 39.3% of the randomized patients. Placebo responses are well-known to be high in clinical trials on RLS.27 In this study we tried to exclude potential placebo responders by means of a 2-week placebo run-in period, but were able to identify only 12 patients; for most patients, the placebo response occurred after several weeks of treatment. These 12 placebo responders were therefore the minority of total placebo responders, and were excluded in an equal proportion from both arms of the study. In our view, the high placebo response rate can be explained by the use of a flexible-dose, parallel design.6,27 Also, in confirmation with previous studies, placebo effects were more evident on subjective rating scales than on PSG studies (table 2).27
Dopamine agonists have been extensively studied in RLS over the last years, but comparative head-to-head studies are not yet available. All of the studies have shown efficacy as well as a placebo response. Taking this into account, the short-term efficacy of pregabalin to treat sensory and motor symptoms was in the same range as dopaminergics. On the other hand, the efficacy of pregabalin (and of other ∝2-δ agonists) to improve sleep architecture can be considered superior to any of the dopamine agonists,6,8,10,28,29 as none of these agents increases SWS.
In general, pregabalin was well-tolerated. Two AEs, unsteadiness and somnolence, were common but mild at a dose of 300 mg/day. Their frequency and severity also showed a dose-dependent pattern, probably related to the daytime administration of doses over 450 mg/day. Although these AEs were generally not problematic in our study population, they could be, under certain circumstances, for part of the working population. However, this will have to be examined more carefully in future studies and will likely depend on what is considered to be the optimal dose. In any case, when used to treat RLS, ∝2-δ compounds are generally administered at night, which partially improves their tolerability.
Despite the fact that brain imaging functional studies have not found consistent evidence for dopaminergic dysfunction in RLS and that there is a lack of consistent support from other studies, dopaminergic dysfunction has been considered to play a central role in the pathophysiology of RLS.30 We believe that the emergence of drugs such as ∝2-δ agonists with similar efficacy to DAs does not falsify the dopaminergic hypothesis but questions its central role. Furthermore, these data may support the hypothesis that spinal reflexes are altered in RLS but modulated by dopamine in the dorsal horn.31 Pregabalin binds to the ∝2-δ receptor of the calcium channel causing a relative reduction in calcium influx into the presynaptic terminals and ultimately, a relative reduction in excitatory release with sparing of inhibitory neurotransmission.32
The duration of our study was insufficient to evaluate augmentation,33 which has been observed under all DAs and, with the exception of tramadol,34,35 is considered to be class-specific. Nevertheless, future studies will need to evaluate pregabalin over the long term. If these studies confirm the combination of a similar short-term efficacy on neurosensitive and motor symptoms to DAs but superior long-term efficacy and effects on sleep, along with an acceptable AE profile, then ∝2-δ agonists will certainly become a therapeutic alternative for RLS.
AUTHOR CONTRIBUTIONS
Statistical analysis was conducted by Dr. Cristina Fernandez.
DISCLOSURE
Dr. Garcia-Borreguero has served on scientific advisory boards for and received speaker honoraria from Boehringer Ingelheim, GlaxoSmithKline, Lundbeck, UCB, Pfizer Inc., XenoPort, Inc., Sanofi-Aventis, Jazz Pharmaceuticals, and Merck & Co., Inc. Dr. Albares has received speaker honoraria from UCB and Sanofi-Aventis. Dr. Larrosa, A.-M. Williams, Dr. Pascual, Dr. Palacios, and Dr. Fernandez report no disclosures.
Footnotes
-
Supplemental data at www.neurology.org
e-Pub ahead of print on April 28, 2010, at www.neurology.org.
Study funding: Supported by an investigator-initiated grant (2005-0408) from Pfizer Inc. (D.G.-B.).
Disclosure: Author disclosures are provided at the end of the article.
Received July 6, 2009. Accepted in final form March 1, 2010.
REFERENCES
Letters: Rapid online correspondence
- Treatment of restless legs syndrome with pregabalin: A double-blind, placebo-controlled study
- Reply from the authors
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
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.
You May Also be Interested in
Hastening the Diagnosis of Amyotrophic Lateral Sclerosis
Dr. Brian Callaghan and Dr. Kellen Quigg