Is there a polysomnographic signature of augmentation in restless legs syndrome?

Sleep Medicine 15 (2014) 1231–1240

Contents lists available at ScienceDirect

Sleep Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / s l e e p

Original Article

Is there a polysomnographic signature of augmentation in restless legs syndrome? Thomas Mitterling, Birgit Frauscher, Tina Falkenstetter, Viola Gschliesser, Laura Ehrmann, David Gabelia, Elisabeth Brandauer, Werner Poewe, Birgit Högl * Department of Neurology, Innsbruck Medical University, Innsbruck, Austria

A R T I C L E

I N F O

Article history: Received 15 January 2014 Received in revised form 8 May 2014 Accepted 14 May 2014 Available online 2 July 2014 Keywords: Sleep Movement disorders Restless legs syndrome therapy Circadian Periodicity Electromyography (EMG) Muscle Recruitment

A B S T R A C T

Objective: Augmentation of restless legs syndrome (RLS) is a potentially severe side-effect of dopaminergic treatment. Data on objective motor characteristics in augmentation are scarce. The aim of this study was to investigate in detail different variables of leg movements (LM) in untreated, treated, and augmented RLS patients. Methods: Forty-five patients with idiopathic RLS [15 untreated, 15 treated (non-augmented), 15 augmented] underwent RLS severity assessment, one night of video-polysomnography with extended electromyographic montage, and a suggested immobilization test (SIT). Results: Standard LM parameters as well as periodicity index (PI) and muscle recruitment pattern did not differ between the three groups. The ultradian distribution of periodic leg movements (PLM) in sleep during the night revealed significant differences only during the second hour of sleep (P < 0.05). However, augmented patients scored highest on RLS severity scales (P < 0.05) and were the only group with a substantial number of PLM during the SIT. Conclusion: This study demonstrates that polysomnography is of limited usefulness for the diagnosis and evaluation of RLS augmentation. In contrast, the SIT showed borderline differences in PLM, and differences on subjective scales were marked. According to these results, augmentation of RLS is a phenomenon that predominantly manifests in wakefulness. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Restless legs syndrome (RLS) is a common sensorimotor disorder characterized by an urge to move the legs, accompanied by unpleasant sensations in the legs, and occurring predominantly during periods of rest in the evening or night [1]. RLS affects ~10% of the general population [2] and in 2.7% of the population the disorder has a moderate-to-severe negative health impact [3]. Levodopa and dopamine agonists are efficacious for the treatment of RLS, but carry a relevant risk of causing augmentation on the long term [4–7]. Augmentation is a worsening of RLS symptom severity during RLS treatment. Features of augmentation include an earlier onset of symptoms, a shorter latency to symptoms at rest, a spread of symptoms to other body parts (e.g. the arms), a shorter duration of the treatment effect, and a paradoxical response to changes in medication (i.e. increase of symptom severity after an increase of the daily medication dosage, decrease of symptom severity after a dose de-

crease) [4,8]. Recent studies have shown that augmentation is present in 11.7% in a clinical series of RLS [9] and a community-based study showed that up to 21% of RLS patients under dopaminergic treatment suffer from augmentation [5]. The pathophysiology of augmentation is not fully understood. Current concepts consider augmentation to reflect a hyperdopaminergic state, in which pronociceptive D1 receptors are believed to be stimulated to a greater extent than antinociceptive D2 receptors, which may generate pain as well as periodic leg movements (PLM) [10]. It is assumed that PLM and motor activity in general increase during augmentation [8]; however, there has been no systematically controlled videopolysomnographic (vPSG) study focused on leg muscle activation in augmentation. The present study aimed to perform a controlled analysis of the characteristics of motor activity in RLS augmentation by vPSG and suggested immobilization test (SIT). 2. Methods

* Corresponding author at: Innsbruck Medical University, Department of Neurology, Anichstrasse 35, A-6020 Innsbruck, Austria. Tel.: +43 512 504 23811; fax: +43 512 504 23842. E-mail address: [email protected] (B. Högl). http://dx.doi.org/10.1016/j.sleep.2014.05.023 1389-9457/© 2014 Elsevier B.V. All rights reserved.

2.1. Study population and design For this study, RLS patients according to standard criteria [1] were prospectively examined at the sleep laboratory of Innsbruck Medical

1232

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

University. Only patients with idiopathic or familiar RLS after exclusion of possible RLS mimics, as proposed by Hening et al. [11], were eligible. Three groups of RLS patients were included: (1) untreated RLS, (2) treated RLS without current augmentation, and (3) treated RLS with current augmentation. In the following, the terms ‘untreated’, ‘treated (non-augmented)’ and ‘augmented’ are used for these groups. The presence of augmentation was determined in a clinical interview according to current standard criteria [8]. Any other disease or medication that might have an impact on motor activity during sleep [e.g. rapid eye movement sleep behaviour disorder or relevant untreated sleep apnea syndrome (apnea hypopnea index (AHI) > 10/h)] represented exclusion criteria. The recruitment aim set for this study was 15 patients in each group. Evaluation of subjects included clinical history, demographic data, neurological examination, RLS-specific scales and questionnaires (see below), a SIT and vPSG. The study had been approved by the ethical committee of Innsbruck Medical University and all participants gave written informed consent.

2.2. RLS scales and questionnaires The International RLS Study Group Rating Scale (IRLS, total score) [12], the RLS-6 Scales (RLS-6) [13], and the Clinical Global Impression (CGI, item 1) [14] were administered to assess severity of RLS symptoms. To analyse the involved body parts, item 4 of the Augmentation Severity Rating Scale (ASRS) [15] was applied.

2.3. Polysomnography with multiple electromyography recording Every patient underwent one night of vPSG. The recording included electroencephalography (EEG montage in accordance with the 2007 American Academy of Sleep Medicine criteria) [16], electrooculography (EOG), respiration using nasal airflow (thermocouple and nasal pressure cannula), and thoracic and abdominal respiratory effort, and one-channel electrocardiography (ECG). In addition to conventional PSG electromyography (EMG) of chin and bilateral tibialis anterior muscles, multichannel EMG of upper and lower extremity muscles (bilateral biceps bracchii, triceps bracchii, rectus femoris, biceps femoris and gastrocnemius muscles) was recorded. EMG signals were recorded with a sampling rate of up to 1000 Hz, high-pass filtered at 50 Hz and low-pass filtered at 300 Hz. Baseline EMG amplitude of the relaxed tibialis anterior muscle was ±2 μV (non-rectified signal). Leg movements (LM) and PLM were recorded according to criteria of the World Association of Sleep Medicine (WASM) [17]. EMG of both tibialis anterior muscles was recorded using surface electrodes placed symmetrically on the middle of the muscles with an inter-electrode distance of 2–3 cm. Sleep stages were scored in 30 s epochs using standard criteria [16].

2.4. Analysis of PLM and related variables In accordance with the aim of this study, the main focus was the analysis of LM and PLM, which was performed in several established and new ways. LM and PLM were analysed manually according to WASM criteria [17]. The LM and PLM indices were calculated for time in bed (TIB), total sleep time (TST), total wake time (TWT), and hour of sleep.

2.4.1. Periodicity index (PI) The PI is calculated as the number of LM intermovement intervals (IMI) occurring in a sequence of three IMIs with a duration between 10 and 90 divided by the total number of IMIs [18].

Table 1 Demographic variables. Untreated (n = 15) Age (median/range) (years) Gender (F/M) RLS duration (median/range) (years) Medication Polytherapy Substances Levodopa Dopamine agonist Pramipexole Ropinirole Rotigotine Opioids Tramadol α2δ Ligands Gabapentin

Treated (non- Augmented P-value (n = 15) augmented) (n = 15)

60 (21–73) 58 (26–72)

60 (38–75)

0.471

10/5 9 (1–25)

10/5 20 (1–61)

0.056 0.050

9/6 16 (4–37)

NA

0

9

NA

1

10

NA NA NA

10 4 0

5 2 1

NA

0

1

NA

0

1

RLS, restless legs syndrome; NA, not applicable.

2.4.2. Muscle recruitment in periodic leg movements in sleep (PLMS) In addition to the standard LM and PLM analysis, every single PLMS was analysed with regard to: (i) the number of involved muscles, and (ii) identification of the first contracting muscle. For these issues, surface EMG of bilateral biceps bracchii, triceps bracchii, rectus femoris, biceps femoris, and gastrocnemius muscles was recorded in addition to the standard tibialis anterior muscle. We calculated the ‘PLMS spread index’, which reflects the proportion of PLMS in which at least two more leg muscles than the tibialis anterior are contracted. This index can vary between 0 (in every PLM less than two leg muscles other than the tibialis contracted) to 1 (in every PLM at least two additional leg muscles are contracted). 2.5. Suggested immobilization test A SIT with a duration of 60 min was conducted before every sleep study. In variation from Michaud et al. [19]. subjects were sitting in a reclining chair in a comfortable position with flected legs in a semi-flexed position. The instructions before the test were to keep the voluntary movements to a minimum for the entire test. Every 15 min the patients were asked to mark their urge to move and leg discomfort on a visual analogue scale (VAS). For each patient the SIT total PLM index is reported as well as the total value (sum of each 15 min value divided by five) on the VAS for the categories ‘urge to move’ and ‘leg discomfort’ and the value of every single 15 min period. To illustrate the evolution of symptoms, values of VAS and PLM were plotted across a time axis. For patients who could not complete the SIT the last observation on VAS carried forward was used for analysis; for calculation of the PLM index for patients prematurely ending the test, the number of PLM was calculated for a 60 min period (i.e. number of PLM divided by number of minutes finished multiplied by 60). 2.6. Statistical analysis Statistical analyses were performed using SPSS 20 (IBM Corp., Armonk, NY, USA). Data are given as frequencies, mean ± standard deviation (SD), median and range, or as mode. Normality testing was performed with the Shapiro–Wilk test. Group comparisons were performed using the χ2-test for categorical variables, and analysis of variance (ANOVA) or Kruskall–Wallis test for quantitative variables depending on distribution. P < 0.05 was considered significant.

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

1233

Table 2 Restless legs syndrome (RLS)-specific scales. Item

(1) Untreated (n = 15)

(2) Treated (non-augmented) (n = 15)

(3) Augmented (n = 15)

P

(1) vs (2)

(1) vs (3)

(2) vs (3)

IRLS RLS-6 item 1. Sleep satisfaction 2. RLS at sleep onset 3. RLS during night 4. RLS during day calm 5. RLS during day moving 6. Sleepiness during day CGI item 1 ASRS item 4

19 (0–33)

11 (0–32)

29 (18–39)

<0.001a

0.148

<0.001a

<0.001a

5 (0–10) 6 (0–10) 4 (0–9) 2 (0–10) 0 (0–1) 3 (0–7) 4 (1–5) 2 (0–4)

4 (0–10) 1 (0–8) 1 (0–5) 3 (0–8) 0 (0–2) 1 (0–7) 3 (2–5) 1 (0–3)

8 (0–10) 7 (0–10) 7 (0–10) 7 (5–10) 1 (0–5) 7 (0–10) 6 (4–6) 3 (1–4)

0.007b 0.003c <0.001d <0.001e 0.011f 0.015g <0.001h 0.015i

0567 0.008 0.056 0.902 0.967 0.285 0.413 0.623

0.010 0.305 0.016 0.002d 0.023 0.05 0.003e 0.019

0.004b 0.002c <0.001d <0.001e 0.037 0.007f <0.001h 0.025

IRLS, International Restless Legs Syndrome Study Group rating scale for RLS; CGI, Clinical Global Impression; ASRS, Augmentation Severity Rating Scale. Values are shown as median and range (IRLS, RLS-6), or as mode and range (CGI 1, ASRS item 4). All P-values significant after correction for multiple comparison: a P < 0.0056; b P < 0.0125; c P < 0.01; d P < 0.0063; e P < 0.0071; f P < 0.016; g P < 0.025; h P < 0.0083; i P < 0.05.

For correction of multiple comparisons, the Bonferroni–Holm method [20] was applied. 3. Results 3.1. Demographic variables Forty-five patients participated in the study (age 54.73 ± 14.98 years; range, 21–75; 29 women, 16 men): 15 patients untreated, 15 treated (non-augmented), and 15 augmented patients. Demographic variables, RLS history, and RLS-specific therapy are shown in Table 1. There was a trend towards longer disease duration in augmented RLS patients (P = 0.05). Other demographic variables did not differ between the three groups.

3.2. RLS-specific treatment and co-medication Untreated patients were either de novo or off RLS-specific treatment for at least 2 weeks prior to the PSG recording. Treated (nonaugmented) subjects had a stable therapy regime for at least 1 week prior to the PSG. Augmented patients underwent polysomnography, taking the medication on which augmentation occurred. Four of these had been prescribed tramadol on demand in the week before the polysomnography in addition to their usual RLS medication. Only one of them was taking tramadol on a regular basis. A combination of RLS-specific drugs was more common in augmented patients compared to treated (non-augmented) subjects. Whereas in this group nine of the 15 patients had a combination of two or more drugs, all non-augmented patients were on

p < 0.001

<0.001 0.148

40

<0.001

IRLS severity scale

30

20

10

0 Untreated

Treated (non-augmented)

Augmented

Fig. 1. Scatterplots of the International Restless Legs Syndrome Study Group rating scale (IRLS) scores. Circles represent single patient values; horizontal lines represent group median values.

1234 T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

Fig. 2. Scatterplot of leg movements (LM) and periodic leg movements (PLM) index. Circles represent single patient values; horizontal lines represent group median values. Outliers are not shown.

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

1235

Fig. 3. Distribution of PLM intermovement intervals (IMI) in sleep (left) and wakefulness (right). The bars show mean values of number of periodic leg movements (PLM), intermovement intervals (IMI) for untreated (upper panels), treated (non-augmented) (middle panels), and augmented (lower panels) patients, whiskers showing standard error of mean.

monotherapy. Details on the substances used in both groups are shown in Table 1. Regarding concomitant medication, only drugs that could have had an effect on sleep, RLS or PLM are reported (hypnotics, antidepressants, antipsychotics). One augmented patient was taking zolpidem; all other subjects were free of medication with effect on sleep, RLS, or PLM. 3.3. Ferritin levels and iron substitution Ferritin levels were available for 39 patients [11 untreated, 13 treated (non-augmented), 15 augmented]. Untreated patients had 97 μg/L (9–213 μg/L), treated (non-augmented) patients had 81 μg/L (7–331 μg/L) and augmented patients had 72 μg/L (5–198 μg/L) [median (range)]. No significant differences were found between the groups. Three patients in each group had a ferritin level <50 μg/L; nine subjects were on regular iron substitution at the time of blood draw [two untreated, three treated (non-augmented), four augmented], three of whom had a ferritin level <50 μg/L (one in every group). 3.4. RLS scales The results of RLS-specific severity scales are shown in Table 2. The patients with current augmentation showed significantly higher values on the IRLS score than non-augmented patients (P < 0.001). Untreated patients had highly variable scores on this scale (range, 0–33). Treated (non-augmented) patients had the lowest scores of all three groups: seven patients had no or mild RLS symptoms and six moderate RLS symptoms. The augmented patients were homogeneous with 13 out of 15 patients with severe to very severe symptoms according to the IRLSSG severity scale (Fig. 1). The CGI item 1 demonstrated significantly higher values for augmented patients compared to the other groups (P < 0.05; Table 2). The RLS-6 scales showed significant differences between groups. Augmented patients scored higher than untreated patients on item 4. In comparison to treated (non-augmented) subjects, augmented patients scored higher on items 1, 2, 3, 4, and 6 (Table 2). Regarding the ASRS

item 4, which reflects the involved body parts in RLS symptoms, augmented patients showed involvement of more body parts than patients of the other groups, yet this did not withstand correction for multiple comparisons (Table 2). 3.5. PSG findings General sleep parameters did not differ between the three groups (online supplementary material). 3.5.1. LM and PLM measures No significant differences were found in the LM and PLM indices between the three groups for time in bed [median, range: untreated vs treated (non-augmented) vs augmented: 31.4, 1.6–95.8 vs 12.3, 4.5–60.9 vs 29.7, 1.5–228.0] or during sleep [median, range: untreated vs treated (non-augmented) vs augmented: 20.6, 0.0– 100.4 vs 6.5, 0.8–57.3 vs 9.9, 0.1–43.0] or during wakefulness [median, range: untreated vs treated (non-augmented) vs augmented: 70.2, 1.8–89.6 vs 36.2, 7.9–155.7 vs 34.8, 7.2–259.4] (Fig. 2). 3.5.2. Distribution of intermovement intervals in PLM The distribution of PLM IMI in sleep and wakefulness is shown in Fig. 3. Visual inspection of distribution of PLM IMI revealed the classical peak of IMI in untreated RLS patients, yet this did not withstand statistical testing when comparing with the other groups. In contrast, PLMW in augmented patients showed a high peak at shortlasting IMI, although there were no significant differences between the groups (Fig. 3). 3.5.3. Calculations of periodicity A scatterplot of the periodicity index across sleep stages is shown in Fig. 4. No significant differences between the groups were found for the periodicity during time in bed, during sleep, and during wakefulness. 3.5.4. Temporal distribution of PLMS The temporal distribution of PLMS is shown in Fig. 5. Across the whole night, significant differences between the groups were found

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

for the PLMS index of the second (P = 0.001) hour of sleep; posthoc analysis revealed a significantly lower value for the second hour of sleep for augmented patients compared with untreated patients (P < 0.001).

Fig. 4. Scatterplot of the periodicity index. Circles represent single patient values; horizontal lines represent group median values.

1236

3.5.5. Muscle recruitment in PLMS For analysis of the recruitment pattern in PLMS, we investigated all PLMS of the 45 patients with regard to (1) which muscle was the first to contract, (2) how many muscles were involved (PLM spread index), (3) which muscles were involved, and (4) the interval from the first to the last activated muscle (from onset to onset). In all groups, the tibialis muscle was by far most often the first contracting muscle during a PLMS [untreated, 89.73%; treated (nonaugmented), 75.32%; augmented, 92.03%]. Calculation of the PLMS spread index, which reflects the proportion of PLMS in which at least two more leg muscles in addition to the tibialis alone are contracted, showed no significant differences between the groups [median, range: untreated, 0.42, 0.08–0.85; treated (not augmented), 0.42, 0.09–0.92; augmented, 0.64, 0.0–1.0; P = 0.595]. A scatterplot of the index is shown in Fig. 6 (right panel). Figure 6 (left panel) shows a graph of the percentage of muscle involvement in PLMS for each group. As expected, these graphs show that, in PLMS, muscles of the lower extremity are more frequently involved than muscles of the upper extremities. The duration of recruitment from the first to the last activated muscle did not significantly differ between the groups [untreated 1.25 (0.82–3.20) vs treated (non-augmented) 1.20 (0.63–2.16) vs augmented 1.08 (0.48–2.88) P = 0.868]. 3.5.6. Suggested immobilization test The SIT was performed before the beginning of the PSG recording between 19:40 and 23:13. Treated subjects were allowed to take their usual RLS-specific medication according to their usual schedule during the SIT. Forty-one of 45 patients finished the test, three patients stopped prematurely due to irresistible urge to move the legs, and in one patient the test was stopped because of sleep attacks. During the SIT, only augmented RLS patients showed a substantial number of PLM, but this marginal difference did not withstand statistical testing: untreated median, 0 (range, 0–191); treated (non-augmented) median, 0 (range, 0–175); augmented median, 62 (range, 0–178) (P = 0.055). Figure 7 shows a scatterplot of the PLM index during the SIT. Regarding the evolution of PLM during the SIT, augmented patients seemed to have more PLM than the other groups in the first two quarters of the test, yet this did not withstand statistical testing (P > 0.0033, P > 0.0036, respectively). Figure 8 shows a graph on evolution of PLM, VAS of urge to move, and VAS of leg discomfort during the SIT. Overall values of the VAS of urge to move and VAS of leg discomfort showed no difference between the groups (P = 0.348, P = 0.210, respectively). Furthermore no significant differences were found in the evolution of VAS of urge to move and VAS of leg discomfort. 4. Discussion The main focus and novel approach of this study consists in the extensive investigation of leg movements and derived measures in RLS by comparing untreated, treated (non-augmented), and augmented patients. The key finding is that standard and more sophisticated leg movement variables in PSG have limited value for the diagnosis and evaluation of RLS augmentation, which could thus be interpreted predominantly as a phenomenon of wakefulness. Accordingly, the SIT was best suitable to discriminate between patients with and without RLS augmentation.

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

1237

Fig. 5. Comparison of periodic leg movements in sleep (PLMS) per hour of sleep between untreated (upper panel), treated (non-augmented) (middle panel) and augmented (lower panel) restless legs syndrome (RLS) patients; bars show mean value, whiskers show standard error of mean.

Fig. 6. (Left) Median percentage of involved muscles in PLMS, whiskers showing 95% confidence interval. (Right) Distribution of PLMS spread index. M.b.b.l., left biceps brachii; M.b.b.r., right biceps brachii; M.t.b.l., left triceps brachii; M.t.b.r., right triceps brachii; M.r.f.l., left rectus femoris; M.r.f.r., right rectus femoris; M.b.f.l., left biceps femoris; M.b.f.r., right biceps femoris; M.g.l., left gastrocnemius; M.g.r., right gastrocnemius; M.t.l., left tibialis anterior; M.t.r., right tibialis anterior; PLMS, periodic leg movements in sleep.

1238

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

p=0.055

0.061 1.0

200

0.037

SIT PLM index

150

100

50

0

Untreated

Treated (non-augmented)

Augmented

Fig. 7. Suggested immobilization test (SIT) periodic leg movements (PLM) index. Circles represent single patient values; horizontal lines represent group median values.

4.1. LM and PLM measures Our study found no differences for standard PLM measures between untreated, treated (non-augmented), and augmented patients. Unexpectedly, patients with augmentation had a relatively low median PLMS index. This is in contrast to a previous study indicating that RLS severity and the PLM index are closely correlated [21], and could perhaps indicate that dopaminergic agents still act on PLM even in augmented patients. Also the periodicity index as calculated by Ferri et al. [18] did not differentiate between augmented and non-augmented patients. However, when analysing the PLMS index for every hour of sleep separately, augmented patients had a lower PLMS index in the second hour of sleep than did untreated patients. This is different from the typical distribution of PLMS per hour of sleep seen in untreated RLS patients, with a maximum value in the first hours of the night [22]. This could indicate a phase delay in augmented patients but a study of circadian rhythms was not within the scope of the present study. Although augmented patients reported a wider bodily distribution of RLS symptoms with subjectively more involved body parts, this was not depicted in the PLM spread index, which reflects the proportion of PLMS in which at least two more leg muscles in addition to the tibialis muscle alone are contracted. This specific additional analysis did not show recruitment of more muscles of the lower or upper extremities during PLM in augmented patients. The overall pattern of muscle recruitment is in line with previous data showing that, in PLMS, predominantly muscles of the lower extremity are involved and that the tibialis anterior is the most frequently recorded starting muscle of PLMS [23]. Interestingly in the treated (non-augmented) group the PLMS index varied largely from 0.8 to 57.3/h. A previous study has found subtle signs of sub-threshold augmentation during the course of stable treatment in at least half of the patients on dopaminergic

treatment with ropinirole, pramipexole, and levodopa [5]. Accordingly one might speculate that the higher PLM indices during stable treatment in some of our patients could point to an early yet subclinical development of tolerance or augmentation. However, longterm studies are needed to corroborate this hypothesis.

4.2. Suggested immobilization test This study suggests that the SIT may contribute to the diagnosis of RLS augmentation. Although all three groups showed a continuous increase on the VAS of urge to move and leg discomfort, solely augmented patients had a substantial number of PLM during the test. This indicates that patients with augmentation cannot suppress their leg movements despite the classical SIT instruction to keep the voluntary movements to a minimum.

4.3. Subjective scales Subjective symptom severity was worst in augmented RLS patients. This is in line with previous reports showing that augmented patients score highest on RLS-specific scales [9]. When plotting the IRLS scores of all three patient groups, the IRLS data form a U-shaped curve (see Fig. 1) which reflects the theory of the dopamine paradox in RLS: an exaggeration of RLS symptoms caused by increased dopamine concentration in augmented patients; a relief from RLS symptoms in optimally treated patients due to dopamine concentration within the normal range; occurrence of RLS symptoms in untreated patients due to a reduced dopamine concentration [10]. Further confirmation of increased daytime symptoms in RLS augmentation is shown by the RLS-6 scores. On item 4, which reflects RLS daytime symptoms at rest, augmented patients scored significantly higher than the other groups.

Fig. 8. Evolution of periodic leg movements (PLM) and subjective symptoms during the suggested immobilization test. Graphs show median number of PLM (left), median score on visual analogue scale (VAS) for leg discomfort (middle) and median score on VAS for urge to move (right) for untreated, treated (non-augmented), and augmented patients.

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

1239

4.4. Ferritin levels Ferritin levels did not differ significantly between augmented and non-augmented patients. This is in contrast to previous reports, which indicated lower ferritin levels in augmented RLS patients [9,24]. One possible explanation for this discrepancy could be the current iron substitution in nine of our patients [four augmented, three treated (non-augmented), two untreated]. We do not know whether iron substitution could have influenced the outcome of the motor analysis. 4.5. Strengths and limitations The major strength of the current work is that it was a thoroughly conducted controlled study with extensive manual characterization of LM and its various derived measures in RLS augmentation, comparing clearly defined patient groups without and with treatment. One potential limitation concerns the sample size of 45 subjects, which was relatively low, and minor differences might thus have been rejected. In addition, the well-established night-to-night variability of PLM [25–27] could at least partially explain the comparatively low median PLM index in augmented RLS. The night-to-night variation of PLM, however, has never been assessed in RLS augmentation, and might be even higher than in nonaugmented patients. Therefore long-term registration of PLM with feasible methodology such as actigraphy might be even more useful to detect augmentation than a single night of PSG. In summary, this study shows that PSG is of limited value in the diagnosis and evaluation of RLS augmentation. However, the SIT was useful to distinguish between augmented and non-augmented RLS, and subjective scales were most sensitive for capturing the symptom burden of RLS in augmented patients. RLS augmentation seems to be a phenomenon that manifests predominantly during wakefulness. Funding source This study was supported by the National Bank of Austria Anniversary fund (Project 12594) to Birgit Högl. Conflicts of interest None declared. The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2014.05.023. Acknowledgements We thank Heinz Hackner for professional support in technical matters. References [1] Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisir J. Restless Legs Syndrome Diagnosis and Epidemiology workshop at the National Institutes of Health; International Restless Legs Syndrome Study Group. Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from the restless legs syndrome diagnosis and epidemiology workshop at the National Institutes of Health. Sleep Med 2003;4:101–19. [2] Högl B, Kiechl S, Willeit J, Saletu M, Frauscher B, Seppi K, et al. Restless legs syndrome: a community-based study of prevalence, severity, and risk factors. Neurology 2005;64:1920–4. [3] Allen RP, Stillman P, Myers AJ. Physician-diagnosed restless legs syndrome in a large sample of primary medical care patients in western Europe: prevalence and characteristics. Sleep Med 2010;11:31–7. [4] Allen RP, Earley CJ. Augmentation of the restless legs syndrome with carbidopa/ levodopa. Sleep 1996;19:205–13.

1240

T. Mitterling et al./Sleep Medicine 15 (2014) 1231–1240

[5] Allen RP, Ondo WG, Ball E, Calloway MO, Manjunath R, Hiqbie RL, et al. Restless legs syndrome (RLS) augmentation associated with dopamine agonist and levodopa usage in a community sample. Sleep Med 2011;12:431–9. [6] Silver N, Allen RP, Senerth J, Earley CJ. A 10-year, longitudinal assessment of dopamine agonists and methadone in the treatment of restless legs syndrome. Sleep Med 2011;12:440–4. [7] Högl B, Garcia-Borreguero D, Trenkwalder C, Ferini-Strambi L, Hening W, Poewe W, et al. Efficacy and augmentation during 6 months of double-blind pramipexole for restless legs syndrome. Sleep Med 2011;12:351–60. [8] García-Borreguero D, Allen RP, Kohnen R, Högl B, Trenkwalder C, Oertel W, et al. Diagnostic standards for dopaminergic augmentation of restless legs syndrome: report from a World Association of Sleep Medicine–International Restless Legs Syndrome Study Group consensus conference at the Max Planck Institute. Sleep Med 2007;8:520–30. [9] Frauscher B, Gschliesser V, Brandauer E, El-Demerdash E, Kaneider M, Rücker L, et al. The severity range of restless legs syndrome (RLS) and augmentation in a prospective patient cohort: association with ferritin levels. Sleep Med 2009;10:611–15. [10] Paulus W, Trenkwalder C. Less is more: pathophysiology of dopaminergictherapy-related augmentation in restless legs syndrome. Lancet Neurol 2006;5:878–86. [11] Hening WA, Allen RP, Washburn M, Lesage SR, Earley CJ. The four diagnostic criteria for restless legs syndrome are unable to exclude confounding conditions (“mimics”). Sleep Med 2009;10:976–81. [12] Walters AS, LeBrocq C, Dhar A, Hening W, Rosen R, Allen RP, et al. Validation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med 2003;4:121–32. [13] Kohnen R, Oertel WH, Stiasny-Kolster K, Benes H, Trenkwalder C. Severity rating of restless legs syndrome: validation of the RLS-6 scales. Sleep 2004;27:A304. [14] National Institute of Mental Health (NIMH). Early clinical drug evaluation unit (ECDEU). Clinical global impressions. In: Guy W, editor. ECDEU assessment manual for psychopharmacology, rev. ed. NIMH. Rockville, MD: 1976. [15] García-Borreguero D, Kohnen R, Högl B, Ferini-Strambi L, Hadjigeorgiou GM, Hornyak M, et al. Validation of the Augmentation Severity Rating Scale (ASRS): a multicentric, prospective study with levodopa on restless legs syndrome. Sleep Med 2007;8:455–63.

[16] Iber C, Ancoli-Israel S, Chesson A, Quan SF, for the American Academy of Sleep Medicine. The AASM manual for the scoring of sleep and associated events: rules, terminology, and technical specifications, 1st ed. Westchester, IL: AASM; 2007. [17] Zucconi M, Ferri R, Allen R, Baier PC, Bruni O, Chokroverty S, et al. The official World Association of Sleep Medicine (WASM) standards for recording and scoring periodic leg movements in sleep (PLMS) and wakefulness (PLMW) developed in collaboration with a task force from the International Restless Legs Syndrome Study Group (IRLSSG). Sleep Med 2006;7: 175–83. [18] Ferri R, Zucconi M, Manconi M, Plazzi G, Bruni O, Ferini-Strambi L. New approaches to the study of periodic leg movements during sleep in restless legs syndrome. Sleep 2006;29:759–69. [19] Michaud M, Paquet J, Lavigne G, Desautels A, Montplaisir J. Sleep laboratory diagnosis of restless legs syndrome. Eur Neurol 2002;48:108–13. [20] Holm S. A simple sequentially rejective multiple test procedure. Scand J Statist 1979;6:65–70. [21] Garcia-Borreguero D, Larrosa O, de la Llave Y, Granizo JJ, Allen R. Correlation between rating scales and sleep laboratory measurements in restless legs syndrome. Sleep Med 2004;5:561–5. [22] Trenkwalder C, Hening WA, Walters AS, Campbell SS, Rahman K, Chokroverty S. Circadian rhythm of periodic limb movements and sensory symptoms of restless legs syndrome. Mov Disord 1999;14:102–10. [23] Provini F, Vetrugno R, Meletti S, Plazzi G, Solieri L, Lugaresi E, et al. Motor pattern of periodic limb movements during sleep. Neurology 2001;57:300–4. [24] Trenkwalder C, Högl B, Benes H, Kohnen R. Augmentation in restless legs syndrome is associated with low ferritin. Sleep Med 2008;9:572–4. [25] Sforza E, Haba-Rubio J. Night-to-night variability in periodic leg movements in patients with restless legs syndrome. Sleep Med 2005;6:259–67. [26] Hornyak M, Kopasz M, Feige B, Riemann D, Voderholzer U. Variability of periodic leg movements in various sleep disorders: implications for clinical and pathophysiologic studies. Sleep 2005;28:331–5. [27] Ferri R, Fulda S, Manconi M, Högl B, Ehrmann L, Ferini-Strambi L, et al. Night-to-night variability of periodic leg movements during sleep in restless legs syndrome and periodic limb movement disorder: comparison between the periodicity index and the PLMS index. Sleep Med 2013;14:293–6.