J Neurol DOI 10.1007/s00415-015-7994-y
ORIGINAL COMMUNICATION
Defining the phenotype of restless legs syndrome/Willis–Ekbom disease (RLS/WED): a clinical and polysomnographic study Fabrizio Rinaldi1,2 • Andrea Galbiati2,3 • Sara Marelli2,3 • Maria Cusmai3 • Alessandro Gasperi3 • Alessandro Oldani2 • Marco Zucconi2 • Alessandro Padovani1 • Luigi Ferini Strambi2,3
Received: 8 September 2015 / Revised: 4 December 2015 / Accepted: 9 December 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Clinical features variability between familial and sporadic restless legs syndrome/Willis–Ekbom disease (RLS/WED) has been previously reported. With this retrospective cohort study, we aimed to determine the clinical and polysomnographic characteristics of 400 RLS/WED patients. Patients with familial RLS/WED were significantly younger than sporadic RLS/WED, while clinical and polysomnographic characteristics were similar in both groups. No difference was found for the age-at-onset between idiopathic and secondary RLS/WED. Periodic limb movements (PLM) index and REM sleep time were higher in idiopathic RLS/WED. Time of onset of symptoms was in the evening or at bedtime in 28.04 and 37.80 % of patients, respectively, while in 21.34 % of patients onset was more than 1 h after sleep onset. Impulse control and compulsive behaviours (ICBs) were found in 13.29 % patients on dopamine agonist therapy. Our analyses support the hypothesis that patients with a familial history of RLS/WED may have a genetic component. Nevertheless, the dichotomy between early and late onset disease seems to be less sharp than previously reported. A large proportion of RLS/WED patients can have atypical features, therefore making the diagnosis challenging. Some cases can be missed even when the patient refers to a sleep specialist, as revealed by the partial absence of daytime & Luigi Ferini Strambi [email protected] 1
Neurology Clinic, Spedali Civili di Brescia Hospital, University of Brescia, Brescia, Italy
2
Department of Clinical Neurosciences, Sleep Disorders Center, San Raffaele Scientific Institute, Milan, Italy
3
Faculty of Psychology, Vita-Salute San Raffaele University, Milan, Italy
symptoms, the high comorbidity with insomnia and other sleep complaints and the high percentage of symptoms beginning after sleep onset. This draws attention on the importance of a careful evaluation of the patient, to recognize potentially treatable secondary forms of RLS/WED. Keywords Restless legs syndrome/Willis-Ekbom disease Polysomnography Impulse control and compulsive behaviours Familial RLS/WED Idiopathic RLS/WED Age of onset
Introduction Restless legs syndrome/Willis–Ekbom disease (RLS/ WED) is a disorder characterized by an irresistible urge to move the legs, concomitant with an unpleasant sensation described as tingling, burning, itching, or pain in the lower limbs. Usually movement relieves these feelings, while they increase in severity at rest, in the evening or during the night [1]. In severe cases, sensory and motor symptoms can also affect the upper limbs [2]. Other characteristics that support the diagnosis include an elevated index (number per hour) of periodic limb movements during sleep (PLMS) and during wakefulness (PLMW), a positive family history of RLS/WED and a relief with dopaminergic treatment [1]. Current studies suggest that the prevalence of RLS/ WED is approximately 2–3 % in the adult Caucasian population [1, 3]. It has been found to have a high degree of familiarity, so that a family history of RLS/WED among first-degree relatives is one of the clinical features supporting the diagnosis [1]. Clinical features variability between familial and sporadic cases of RLS/WED has been compared in some of studies; an association between an early age-at-onset and the presence of a positive family
123
J Neurol
history of RLS/WED has been reported [4]. The cut-off values for differentiating ‘‘early’’ from ‘‘late’’ age of onset, however, have been rather arbitrarily defined [5, 6]. Furthermore, these studies that separated RLS/WED patients on the basis of age-at-onset did not report detailed polysomnographic data. The most frequent form of RLS/WED is called ‘‘primary’’ or ‘‘idiopathic’’ (iRLS/WED). Nevertheless, several conditions, such as anemia [7], renal failure [8] and peripheral neuropathy [9] were shown to be associated with RLS/WES, and they are considered ‘‘secondary’’ forms (sRLS/WED). Ekbom [10] first reported a high incidence of iron deficiency among patients with RLS/WED. A significant correlation between subjective ratings of RLS/ WED severity, serum ferritin and sleep efficiency was later described [11, 12]. The aim of the present retrospective study was to determine the clinical and polysomnographic characteristics of a large cohort of patients with idiopathic and secondary forms, with a focus on the distribution of age-atonset overall as well as on familiarity for RLS/WED. We also aimed to test the presence of atypical features, such as late time of symptom onset, and the risk of developing impulse control and compulsive behaviours (ICBs) in relation to dopaminergic treatment.
Methods 400 RLS/WED patients (229 women, mean age 62.05 ± 13.80 years) consecutively assessed at the Sleep Centre of the San Raffaele Hospital in Milan, Italy were enrolled for this study, from 2010 to 2014. All these patients met the criteria established by the International Restless Legs Syndrome Study Group [13]. Diagnosis was based on medical history, with a comprehensive interview for sleep disorder performed by a physician expert in sleep disorder, in addition to physical and neurological exams also carried out by a neurologist trained in sleep medicine and movement disorders. Information including age-at-onset, comorbidities, familial history, time of symptoms onset, symptoms severity, treatment and the presence of ICBs were obtained. RLS/WED patients were classified as familial if they had at least one first-degree relative who also had RLS/WED, or sporadic if they had no first-degree relatives with RLS/ WED. We considered sRLS/WED when the patient was affected by neuropathy, renal failure, amyloidosis and low ferritin values. Symptoms severity was assessed by means of the IRLS total score, while frequency of symptoms was obtained by item 7 of the IRLS [14].
123
The RLS/WED topography patterns were classified according to localization in upper limbs (UL), lower limbs (LL) or both and lateralization as symmetrical or asymmetrical limbs involvement. 122 drug naı¨ve RLS/WED patients underwent a nocturnal polysomnographic (PSG) evaluation, with an adaption night preceding the PSG recording. We considered as drug naı¨ve only the patients who were neither taking medications for RLS/WED nor other drugs that could have affected sleep, such as benzodiazepines or antidepressants. Lights-out time was based on the individual’s usual bedtime and ranged between 11.00 and 11:30 p.m. The following signals were recorded: electroencephalogram (EEG) (six channels, including F4–M1, C4–M1, 02–M1 with backup electrodes at F3, C3, O1 and M2); electrooculogram (EOG); EMG of the submentalis muscle; EMG of the right and left tibialis anterior muscles; electrocardiogram (ECG; one derivation) according to the American Academy of Sleep Medicine (AASM) scoring criteria [15]. The sleep respiratory pattern of each patient was monitored using oral and nasal airflow thermistors and/ or nasal pressure cannula, thoracic and abdominal respiratory effort strain gauge and by monitoring oxygen saturation (pulse-oximetry). Sleep stages were scored following AASM criteria [15] on 30-s epochs. Leg movements (LM) during sleep were detected following WASM criteria [16]. Patients were followed up at 3- to 6-month intervals, and eventually started a pharmacological treatment whenever symptoms conferred adverse effects on sleep, quality of life and health. Statistical analyses were performed using SPSS 13.0. ANOVA was employed to evaluate differences between iRLS/WED and sRLS/WED, and between familial and sporadic disease. When multiple comparisons were performed we employed Bonferroni correction. The Spearman’s rank correlation coefficient was calculated to test possible association between clinical, demographic and PSG data.
Results Clinical phenotype Out of 400 subjects, 229 were female (57.25 %). Duration of the disease was 18.26 ± 15.75 years, with women having a longer mean duration (male 15.62 ± 14.57 vs female 20.23 ± 16.33 years; F = 8.543, p = 0.004). RLS/ WED duration did not have any significant impact on PSG variables. On the other hand, age had a significant correlation with wake after sleep onset (WASO) (r = 0.319; p = 0.000), PLM (r = 0.361; p = 0.000) and REM
J Neurol
latency (r = 0.214; p = 0.000), and an inverse correlation with SE (r = -0.286; p = 0.002) and TST (r = -0.235; p = 0.010). Age of symptoms onset was lower in female (male 47.17 ± 16.60 vs female 43.24 ± 16.11 years; p = 0.019). It had a significant correlation with PLM (r = 0.292; p = 0.002) and REM latency (r = 0.261; p = 0.006), and an inverse correlation with sleep efficiency (SE) (r = -0.189; p = 0.041) and total sleep time (TST) (r = -0.185; p = 0.044). Nevertheless, our analysis showed that the associations found between the age of onset and polysomnographic measures of disrupted sleep are linked to aging itself more than to the age of onset. In particular, regarding SE and PLM, age had a significant effect (SE: p \ 0.005 and PLM: p \ 0.005). This is in line with the fact that RLS duration did not have any impact on PSG variables, while age was related to worse PSG characteristics. Mean scores at IRLS and ESS were 24.02 ± 6.17 and 6.34 ± 4.47, respectively, indicating that patients had a severe form of RLS/WED but without excessive daytime sleepiness. RLS severity as measured by IRLS scores correlated positively with age (r = 0.264; p \ 0.001) and with disease duration (r = 0.297; p \ 0.001). Time of onset of RLS/WED symptoms was in the evening or at bedtime in 28.04 and 37.80 % of patients, respectively. However, 21.34 % of patients reported onset of symptoms more than 1 h after falling asleep. Only 5.48 % of patients complained of symptoms onset in the afternoon, while 7.31 % reported the occurrence of symptoms both at day and at night (see Table 1). 173 patients were treated with at least one medication for RLS/WED. Of these, 89.42 % were on treatment with dopamine agonist (DA) (pramipexole, mean dosage 0.22 ± 0.16 mg; ropinirole, mean dosage 1.13 ± 0.73 mg; rotigotine, mean dosage 2 ± 0 mg), 40.38 % with clonazepam (mean dosage 0.66 ± 0.39 mg), 35.57 % with a-2d ligands (pregabalin, mean dosage 103.34 ± 68.7 mg; gabapentin, mean dosage 345.45 ± 181.86 mg) and 0.96 % with opioids. Most of the patients were on monotherapy (42.34 %), while 39.42 % used two different drugs and 18.24 % used more than two. We compared the PSG characteristics of these treated patients with the drug naı¨ve patients. The treated patients showed a lower PLM index (p \ 0.001), a longer REM latency (p \ 0.005) and a shorter REM percentage (p \ 0.05). More specifically, we found that dopamine agonist had a significant effect in reducing PLM index (p \ 0.05) and REM percentage (p \ 0.05) and in increasing REM sleep latency (p \ 0.05). ICBs were found in 13.29 % RLS/WED patients on DA therapy; most of these were related to abnormal eating (7.51 %) and sexual (4.64 %) behaviours, while gambling and punding were reported in 1.15 and 2.31 %, respectively (see Table 1).
Table 1 Clinical phenotype of RLS/WED patients Age
62.05 ± 13.80
Sex
229 women (57 %) 171 men (43 %)
Time of RLS/WED symptoms onset
Evening 28.04 % Bedtime 37.80 % More than 1 h after sleep onset 21.34 % In the afternoon 5.48 % Both at day and at night 7.31 %
Impulse control symptoms
Total 13.29 % Abnormal eating 7.51 % Sexual behaviours 4.64 % Gambling 1.15 % Punding 2.31 %
Comorbidities
Diabetes 3.93 % Psychiatric disorders 16.72 % Hypertension 13.44 % Neuropathies 4.59 % Sleep difficulties 80 % Uremia 0.32 %
Localization of symptoms
Bilateral and symmetric lower limbs (LL) 74.7 % Asymmetric LL 4.1 % Upper limbs 0.7 % Four limbs symmetrical 17.8 % Asymmetrical 2.7 %
Age-at-onset
44.93 ± 16.42
Considering comorbidities, most of the patients complained of some kind of sleep difficulties (80 %); other frequent diseases were psychiatric disorders (16.72 %), hypertension (13.44 %), neuropathies (4.59 %) and diabetes (3.93 %). Only 0.32 % of patients reported the presence of uremia. In 146 untreated patients, information on the localization of symptoms were collected. Bilateral and symmetric lower limb (LL) location was the most frequent (74.7 %), while 4.1 % of participants exhibited asymmetric LL localization. Only one patient (0.7 %) showed symptoms confined to the upper limbs (UL), symmetrically. Four limbs involvement was found in 20.5 % of patients, and was symmetrical in 86.8 % of cases. Overall, 6.9 % of patients complained of asymmetrical symptoms. The severity of RLS/WED as measured with IRLS was similar across the patients, independent of the localization of symptoms. Conversely, IRLS values correlated positively with PLM (r = 0.205; p = 0.044), N2 (r = 0.219; p = 0.030) and negatively with REM sleep (r = -0.243; p = 0.15). The single regression analyses showed that PLM index (p \ 0.05), duration of disease (p \ 0.001) and ESS values (p \ 0.05) are predictors of IRLS values.
123
J Neurol
We assessed the presence of pregnancy in the personal history of the female patients, and its effects on RLS/WED symptoms. 34.4 % had no pregnancy in their life, while 31.8 % reported no effect, 4.6 % mild worsening, 7.2 % moderate worsening, 6.7 % severe worsening and 15.4 % had no recall of this effect. Familial vs. sporadic RLS/WED Familial RLS/WED was present in 186 (53.5 %) patients, while 214 (46.5 %) had a sporadic form. Mean age-at-onset differed as a function of presence/absence of a familial history of RLS/WED (40.44 ± 16.43 vs. 49.03 ± 15.33 years; p \ 0.001). Familial RLS/WED patients were also younger (60.47 ± 14.54 vs. 63.41 ± 12.96 years; p \ 0.05). These differences were significant even when adjusted for age and gender. Polysomnographic characteristics were similar in both groups [including sleep latency (SL); WASO; SE; TST; number of awakenings: PLM index; stage N1; stage N2; stage N3; REM; REM latency]. Likewise, RLS/WED severity as reported by the IRLS scale was similar between the two groups. The two groups did not differ even in the ESS values and in the serum ferritin values (see Table 2). Table 2 Familial vs. sporadic RLS/WED Familial
Idiopathic vs. secondary RLS/WED 277 out of 400 subjects had iRLS/WED (49.50 % female). No differences were found in the age-at-onset between idiopathic and secondary forms (44.89 ± 16.35 vs. 45.02 ± 16.64 years) or in the age of the patients (62.10 ± 13.83 vs. 61.93 ± 13.74 years). Regarding the PSG findings, PLM index was significantly higher in iRLS/WED (50.07 ± 47.72 vs. 32.39 ± 41.46; p = 0.047), while REM duration was longer in iRLS/WED (20.08 ± 6.22 vs 17.48 ± 6.91 min; p = 0.043). These differences were not significant when adjusted for age and gender. The other PSG parameters did not differ between the two groups. IRLS and ESS values were higher in sRLS/WED (IRLS score: 25.365 ± 6.35 vs 23.33 ± 5.98, p \ 0.05; ESS score 7.43 ± 5.22 vs 5.79 ± 3.93, p \ 0.05), even when adjusted for age and gender, while ferritin values were similar between the two groups (see Table 3).
Discussion The aim of this study was to describe a large cohort of RLS/WED patients, comparing hereditary and sporadic as well as idiopathic and secondary disease. We focused on Table 3 Idiopathic vs. secondary RLS/WED
Sporadic
p
Idiopathic
SL
28.13 ± 37.22
28.13 ± 25.46
NS
SL
WASO
91.57 ± 85.30
114.95 ± 82.63
NS
WASO
SE TST
73.39 ± 17.91
72.67 ± 17.43
NS
SE
360.15 ± 96.84
369.22 ± 92.28
NS
TST
Secondary
p
28.64 ± 38.47
26.84 ± 53.03
NS
102.61 ± 84.06
98.28 ± 86.64
NS
72.45 ± 17.80
74.22 ± 17.55
NS
362.39 ± 94.73
366.34 ± 95.80
NS
AWK
14.29 ± 7.80
14.00 ± 6.53
NS
AWK
14.24 ± 6.61
14.07 ± 8.42
NS
PLMSI
45.30 ± 49.94
40.59 ± 40.07
NS
PLMSI
50.07 ± 47.72
32.39 ± 41.46
NS
Stage N1
5.58 ± 4.08
6.64 ± 6.05
NS
Stage N1
6.19 ± 5.36
5.71 ± 4.29
NS
Stage N2
52.31 ± 12.59
49.58 ± 13.84
NS
Stage N2
50.06 ± 13.46
53.23 ± 12.42
NS
Stage N3
22.87 ± 9.91
25.20 ± 11.61
NS
Stage N3
23.71 ± 11.27
24.03 ± 9.54
NS
Stage R
19.61 ± 5.87
18.49 ± 7.48
NS
Stage R
20.08 ± 6.22
17.48 ± 6.91
NS
130.88 ± 83.97 96.90 ± 119.34
129.93 ± 69.56 115.12 ± 126.66
NS NS
Stage R latency Ferritin
131.37 ± 83.75 160.69 ± 139.82
128.91 ± 67.56 49.30 ± 65.89
NS NS
IRLS
24.36 ± 6.39
23.68 ± 5.95
NS
IRLS
23.33 ± 5.98
25.365 ± 6.35
\0.05§*
ESS
6.16 ± 4.35
6.51 ± 4.58
NS
\0.05§*
Stage R latency Ferritin
Age Age of onset
60.47 ± 14.54 40.44 ± 16.43
63.41 ± 12.96 49.02 ± 15.33
ESS
5.79 ± 3.93
7.43 ± 5.22
§
Age
62.10 ± 13.83
61.93 ± 13.74
NS
§
Age of onset
44.89 ± 16.35
45.02 ± 16.64
NS
\0.05 * \0.05 *
SL sleep latency, WASO wake after sleep onset, SE percent sleep efficiency, TST total sleep time, AWK awakenings, PLMSI periodic limb movements of sleep index, stage N1 stage non-REM 1, stage N2 stage non-REM 2, stage N3 stage non-REM3, stage R stage REM, IRLS international restless legs syndrome rating scale, ESS Epworth Sleepiness Scale § Adjusted for age and gender
SL sleep latency, WASO wake after sleep onset, SE percent sleep efficiency, TST total sleep time, AWK awakenings, PLMSI periodic limb movements of sleep index, stage N1 stage non-REM 1, stage N2 stage non-REM 2, stage N3 stage non-REM3, stage R stage REM, IRLS international restless legs syndrome rating scale, ESS Epworth Sleepiness Scale § Adjusted for age and gender
* Adjusted for Bonferroni correction
* Adjusted for Bonferroni correction
123
J Neurol
the age of onset, the polysomnographic findings and the clinical signs and symptoms of RLS/WED. Our study revealed the presence of several atypical features in a large proportion of patients with RLS/WED, which may pose a diagnostic challenge. Firstly, even though most of the patients reported the onset of symptoms in the evening or at bedtime, a relevant percentage complained of symptom beginning 1 h or more after sleep onset. This is of peculiar importance when investigating the presence of symptoms suggestive of RLS/WED in patients with sleep maintenance insomnia. Secondly, we found a higher percentage of ICBs than reported in most of the previous works (13.29 vs. 10 % in Lipford et al. [35], 8.6 % in Ondo et al. [36], 12.4 % in Pourcher et al. [37], 17 % in Cornelius et al. [38]). This is of relevance considering that, in our cohort of patients, the mean dosage of pramipexole was much lower (0.22 mg, compared to 1.33 mg in the study by Lipford et al.) [35], and sexual behaviours were more frequent than previously described [37, 39, 40]. Our results point to the importance of taking a careful history of risk factors when offering a treatment with DA, even if at low dosage, forewarning the patients of potential side effects and planning a close follow-up. The use of a a-2-d calcium-channel ligand should also be considered [41]. The importance of a careful evaluation is furthermore underlined by the frequent occurrence of sleep complaints, experienced by over 80 % of the patients. In these cases, in fact, RLS/WED symptoms may go unnoticed because of more manifest sleep disorders. Moreover, the age-at-onset distribution in the whole sample of our study did not show the diphasic distribution described in previous works. Defining a cut-off age for distinguishing early onset from late onset RLS/WED has always been intricate, and previous works posed a mark at different ages, varying from 36 years [29] to 45 years [20]. One possible explanation for these differences is that the age for differentiating ‘‘early’’ from ‘‘late’’ had been distorted because of smaller samples. We, therefore, suggest that physicians do not rely solely on age of onset for identifying the etiology of RLS/WED. Nevertheless, a younger onset of RLS/WED symptoms should raise suspicions for a familial form. On the other side, some of our findings are in line with the literature data, and can therefore add information that can support hypotheses about the pathophysiology of RLS/ WED. The severity of RLS/WED as measured with IRLS correlated positively with age and with disease duration, confirming the view of a slow progression of symptoms [20]. Furthermore, notwithstanding the presence of high IRLS scores, mean ESS scores were definitely below the
threshold for a significant daytime sleepiness [1]. This finding supports the hypothesis that RLS/WED enhances alertness despite sleep loss, thanks to hyperarousal [1, 21]. According to this theoretical view, hyperarousal is caused by an increased level of glutamatergic activity in the thalamus. This would lead, on one side, to an increased waking time during sleep, and on the other side, to a decreased daytime somnolence. Hyperarousal can also explain the persistence of insomnia in patients with RLS/ WED otherwise successfully treated with dopamine agonists [22]. Previous surveys differed somewhat regarding gender distribution, with some reporting a higher prevalence for women [23, 24] while others did not [25, 26]. In our sample, 57.25 % of patients were females, supporting the view that female sex is a risk factor for RLS/WED [1]. Female patients also showed a younger age of onset. Talking about age of onset of the disease, our study shows that patients with a familial history of RLS/WED were almost 10 years younger than those without, confirming previous findings [5, 20, 29, 30], and supporting the hypothesis that early onset RLS/WED may have a genetic component. Besides that, we did not find any significant difference in clinical or PSG characteristics between familial and sporadic RLS/WED. We also found a lower percentage of familial RLS/WED than previously reported (53.5 vs up to 60–65 % of cases [6, 27]). These differences between studies may be ascribed to different definitions of ‘‘familial’’. In fact, relatives are not always available for a direct assessment by sleep specialist, and even less for a complete PSG evaluation, so the definition of ‘‘familiarity’’ relies on the physician’s feeling about the patients’ story about their relatives’ symptoms. Since the genetic of RLS/ WED is still not completely clarified [28], we considered a positive family history only if first a grade relative was diagnosed affected by the disease. On the contrary, the age of onset of secondary and idiopathic RLS/WED was similar, although previous works found an earlier onset of symptoms in the latter [29]. Even in this case, an explanation for this discrepancy may lie in the different definition of ‘‘secondary’’ RLS/WED, encompassing association with uremia, rheumatoid arthritis, amyloidosis, iron deficiency, folate and B12 deficiency [11, 31–34]. Regarding our sample, there was a particularly low amount of cases of RLS/WED due to ‘‘end-stage renal disease’’, with a relatively high number of neuropathy-related RLS/WED. This could be explained by the older age of sRLS/WED patients of our sample, or due to a selection bias. The comorbidities shown by our sample were otherwise similar to what were expected from the literature data [17–19] (see Table 1). Regarding the polysomnographic findings, a lower PLM index was found in treated patients, confirming the efficacy
123
J Neurol
of treatment. They also displayed a longer REM latency and a shorter REM stage duration. This may be mainly due to the usage of DA, that can delay REM sleep latency and decrease total percentage of REM sleep [42, 43]. Our study shows some biases. First, being retrospective and not population based, it may be due to selection bias. The sample included either patients with and without DA treatment or other medications that could have influenced in some part of the results. Furthermore, since a significant amount of patients was under a polytherapy we did not perform sub-group analyses to avoid biases due to skewed data in small subgroups of patients. The duration of treatment also was very different among those patients. Moreover, we were not able to interview relatives of patients, and this may have caused erroneous placement into the familial or sporadic groups. On the other hand, this work has some strengths. It displays a large cohort of patients with RLS/WED who were evaluated by the same team of physicians in a relatively short period, with a substantial amount undergoing PSG evaluation, both with and without treatment. The presence of atypical symptoms in a large proportion of patients is of peculiar importance and points to the importance of obtaining a thorough history.
Conclusions RLS/WED is an easy recognizable disease when all diagnostic criteria are met, but a large proportion of patients can have atypical features, therefore making the diagnosis challenging. Some cases can be missed even when the patient refers to a sleep specialist, as revealed by the partial absence of daytime symptoms, the high comorbidity with insomnia and other sleep complaints and the high percentage of symptoms beginning after sleep onset. The dichotomy between early and late onset disease seems to be less sharp than previously reported. This draws attention on the importance of a careful evaluation of the patient, to recognize potentially treatable secondary forms of RLS/WED. The overall similarity of clinical characteristics and the course of the disease in idiopathic, hereditary and sporadic RLS/WED could suggest a common pathological pathway. The risk of developing ICBs seems to occur more frequently and with lower dosages of dopaminergic drugs than previously reported. The physician should be aware of this risk and consider the use of alternative medications. Compliance with ethical standards Conflicts of interest of interest.
123
The authors declare that they have no conflict
Ethical standards All participants provided written informed consent to the experimental procedure, which was previously approved by the local ethical committee.
References 1. Allen RP, Picchietti DL, Garcia-Borreguero D, Ondo WG, Walters AS, Winkelman JW, Zucconi M, Ferri R, Trenkwalder C, Lee HB, International Restless Legs Syndrome Study Group (2014) Restless legs syndrome/Willis-Ekbom disease diagnostic criteria: updated International Restless Legs Syndrome Study Group (IRLSSG) consensus criteria—history, rationale, description, and significance. Sleep Med 15(8):860–873 2. Chabli A, Michaud M, Montplaisir J (2000) Periodic arm movements in patients with the restless legs syndrome. Eur Neurol 44(3):133–138 3. Garcia-Borreguero D, Egatz R, Winkelmann J, Berger K (2006) Epidemiology of restless legs syndrome: the current status. Sleep Med Rev 10(3):153–167 4. Winkelmann J (2002) The genetics of restless legs syndrome. Sleep Med 3(Suppl):S9–S12 5. Winkelmann J, Muller-Myhsok B, Wittchen HU, Hock B, Prager M, Pfister H, Strohle A, Eisensehr I, Dichgans M, Gasser T, Trenkwalder C (2002) Complex segregation analysis of restless legs syndrome provides evidence for an autosomal dominant mode of inheritance in early age at onset families. Ann Neurol 52(3):297–302 6. Allen RP, La Buda MC, Becker P, Earley CJ (2002) Family history study of the restless legs syndrome. Sleep Med 3(Suppl):S3–S7 7. Aul EA, Davis BD, Rodnitzky RL (1998) The importance of formal serum iron studies in the assessment of restless legs syndrome. Neurology 51(3):912 8. Winkelman JW, Chertow GM, Lazarus JM (1996) Restless legs syndrome in end-stage renal disease. Am J Kidney Dis 28(3):372–378 9. Ondo W, Jankovic J (1996) Restless legs syndrome: clinicoetiologic correlates. Neurology 47:1435–1441 10. Ekbom KA (1960) Restless legs syndrome. Neurology 10:868–873 11. O’Keeffe ST, Gavin K, Lavan JN (1994) Iron status and restless legs syndrome in the elderly. Age Ageing 23(3):200–203 12. Sun ER, Chen CA, Ho G, Earley CJ, Allen RP (1998) Iron and the restless legs syndrome. Sleep 21(4):371–377 13. Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisi J, Restless Legs Syndrome Diagnosis and Epidemiology workshop at the National Institutes of Health, International Restless Legs Syndrome Study Group (2003) 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 4(2):101–119 14. Walters AS, LeBrocq C, Dhar A, Hening W, Rosen R, Allen RP, Trenkwalder C, International Restless Legs Syndrome Study Group (2003) Validation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med 4(2):121–132 15. Iber C, Ancoli-Israel S, Chesson AL, Quan SF, For the American Academy of Sleep Medicine (2007) The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications, 1st edn. American Academy of Sleep Medicine, Westchester 16. Zucconi M, Ferri R, Allen R, Baier PC, Bruni O, Chokroverty S, Ferini-Strambi L, Fulda S, Garcia-Borreguero D, Hening WA,
J Neurol
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Hirshkowitz M, Ho¨gl B, Hornyak M, King M, Montagna P, Parrino L, Plazzi G, Terzano MG, International Restless Legs Syndrome Study Group (IRLSSG) (2006) 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 7(2):175–183 Benediktsdottir B, Janson C, Lindberg E, Arnardo´ttir ES, Olafsson I, Cook E, Thorarinsdottir EH, Gislason T (2010) Prevalence of restless legs syndrome among adults in Iceland and Sweden: Lung function, comorbidity, ferritin, biomarkers and quality of life. Sleep Med 11(10):1043–1048 Szentkira´lyi A, Vo¨lzke H, Hoffmann W, Trenkwalder C, Berger K (2014) Multimorbidity and the risk of restless legs syndrome in 2 prospective cohort studies. Neurology 82(22):2026–2033 Ulfberg J, Bjorvatn B, Leissner L, Gyring J, Karlsborg M, Regeur L, Skeidsvoll H, Polo O, Partinen M, Nordic RLS Study Group (2007) Comorbidity in restless legs syndrome among a sample of Swedish adults. Sleep Med 8(7–8):768–772 Allen RP, Earley CJ (2000) Defining the phenotype of the restless legs syndrome (RLS) using age-of-symptom-onset. Sleep Med 1(1):11–19 Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, Nissen C (2010) The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev 14(1):19–31 Ferri R, Cosentino FI, Manconi M, Rundo F, Bruni O, Zucconi M (2014) Increased electroencephalographic high frequencies during the sleep onset period in patients with restless legs syndrome. Sleep 37(8):1375–1381 Phillips B, Hening W, Britz P, Mannino D (2006) Prevalence and correlates of restless legs syndrome: results from the 2005 National Sleep Foundation Poll. Chest 129(1):76–80 Rothdach AJ, Trenkwalder C, Haberstock J, Keil U, Berger K (2000) Prevalence and risk factors of RLS in an elderly population: the MEMO study. Memory and morbidity in Augsburg elderly. Neurology 54(5):1064–1068 Lavigne GJ, Montplaisir JY (1994) Restless legs syndrome and sleep bruxism: prevalence and association among Canadians. Sleep 17(8):739–743 Phillips B, Young T, Finn L, Asher K, Hening WA, Purvis C (2000) Epidemiology of restless legs symptoms in adults. Arch Internal Med 160(14):2137–2141 Bonati M, Ferini-Strambi L, Aridon P, Oldani A, Zucconi M, Casari G (2003) Autosomal dominant restless legs syndrome maps an chromosome 14q. Brain 126(Pt 6):1485–1492 Schormair B, Winkelmann J (2011) Genetics of restless legs syndrome: Mendelian, complex, and everything in between. Sleep Med Clin 6(2):203–215 Whittom S, Dauvilliers Y, Pennestri MH, Vercauteren F, Molinari N, Petit D, Montplaisir J (2007) Age-at-onset in restless legs
30.
31.
32.
33. 34. 35.
36.
37.
38.
39.
40.
41.
42.
43.
syndrome: a clinical and polysomnographic study. Sleep Med 9(1):54–59 Hanson M, Honour M, Singleton A, Crawley A, Singleton A, Hardy J, Gwinn-Hardy K (2004) Analysis of familial and sporadic restless legs syndrome in age of onset, gender, and severity features. J Neurol 251(11):1398–1401 Collado-Seidel V, Kohnen R, Samtleben W, Hillebrand GF, Oertel WH, Trenkwalder C (1998) Clinical and biochemical findings in uremic patients with and without restless legs syndrome. Am J Kidney Dis 31(2):324–328 Reynolds G, Blake DR, Pall HS, Williams A (1986) Restless leg syndrome and rheumatoid arthritis. Br Med J (Clin Res Ed) 292(6521):659–660 Heinze EG, Frame B, Fine G (1967) Restless legs and orthostatic hypotension in primary amyloidosis. Arch Neurol 16(5):497–500 Botez MI, Lambert B (1977) Folate-deficiency and restless-legssyndrome in pregnancy. N Engl J Med 297(12):670 Lipford MC, Silber MH (2012) Long-term use of pramipexole in the management of restless legs syndrome. Sleep Med 13(10):1280–1285 Ondo WG, Lai D (2008) Predictors of impulsivity and reward seeking behavior with dopamine agonists. Parkinsonism Relat Disord 14(1):28–32 Pourcher E, Re´millard S, Cohen H (2010) Compulsive habits in restless legs syndrome patients under dopaminergic treatment. J Neurol Sci 290(1–2):52–56 Cornelius JR, Tippmann-Peikert M, Slocumb NL, Frerichs CF, Silber MH (2010) Impulse control disorders with the use of dopaminergic agents in restless legs syndrome: a case-control study. Sleep 33(1):81–87 Schreglmann SR, Gantenbein AR, Eisele G, Baumann CR (2012) Transdermal rotigotine causes impulse control disorders in patients with restless legs syndrome. Parkinsonism Relat Disord 18(2):207–209 Voon V, Schoerling A, Wenzel S, Ekanayake V, Reiff J, Trenkwalder C, Sixel-Do¨ring F (2011) Frequency of impulse control behaviours associated with dopaminergic therapy in restless legs syndrome. BMC Neurol 11:117 Garcia-Borreguero D, Kohnen R, Silber MH, Winkelman JW, Earley CJ, Ho¨gl B, Manconi M, Montplaisir J, Inoue Y, Allen RP (2013) The long-term treatment of restless legs syndrome/WillisEkbom disease: evidence-based guidelines and clinical consensus best practice guidance: a report from the International Restless Legs Syndrome Study Group. Sleep Med 14(7):675–684 Montplaisir J, Nicolas A, Denesle R et al (1999) Restless legs syndrome improved by pramipexole. A double-blind randomized trial. Neurology 52:938–943 Saletu M, Anderer P, Saletu-Zyhlarz G et al (2002) Acute placebo-controlled sleep laboratory studies and clinical follow-up with pramipexole in restless legs syndrome. Eur Arch Psychiatry Clin Neurosci 252:185–194
123
ORIGINAL COMMUNICATION
Defining the phenotype of restless legs syndrome/Willis–Ekbom disease (RLS/WED): a clinical and polysomnographic study Fabrizio Rinaldi1,2 • Andrea Galbiati2,3 • Sara Marelli2,3 • Maria Cusmai3 • Alessandro Gasperi3 • Alessandro Oldani2 • Marco Zucconi2 • Alessandro Padovani1 • Luigi Ferini Strambi2,3
Received: 8 September 2015 / Revised: 4 December 2015 / Accepted: 9 December 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Clinical features variability between familial and sporadic restless legs syndrome/Willis–Ekbom disease (RLS/WED) has been previously reported. With this retrospective cohort study, we aimed to determine the clinical and polysomnographic characteristics of 400 RLS/WED patients. Patients with familial RLS/WED were significantly younger than sporadic RLS/WED, while clinical and polysomnographic characteristics were similar in both groups. No difference was found for the age-at-onset between idiopathic and secondary RLS/WED. Periodic limb movements (PLM) index and REM sleep time were higher in idiopathic RLS/WED. Time of onset of symptoms was in the evening or at bedtime in 28.04 and 37.80 % of patients, respectively, while in 21.34 % of patients onset was more than 1 h after sleep onset. Impulse control and compulsive behaviours (ICBs) were found in 13.29 % patients on dopamine agonist therapy. Our analyses support the hypothesis that patients with a familial history of RLS/WED may have a genetic component. Nevertheless, the dichotomy between early and late onset disease seems to be less sharp than previously reported. A large proportion of RLS/WED patients can have atypical features, therefore making the diagnosis challenging. Some cases can be missed even when the patient refers to a sleep specialist, as revealed by the partial absence of daytime & Luigi Ferini Strambi [email protected] 1
Neurology Clinic, Spedali Civili di Brescia Hospital, University of Brescia, Brescia, Italy
2
Department of Clinical Neurosciences, Sleep Disorders Center, San Raffaele Scientific Institute, Milan, Italy
3
Faculty of Psychology, Vita-Salute San Raffaele University, Milan, Italy
symptoms, the high comorbidity with insomnia and other sleep complaints and the high percentage of symptoms beginning after sleep onset. This draws attention on the importance of a careful evaluation of the patient, to recognize potentially treatable secondary forms of RLS/WED. Keywords Restless legs syndrome/Willis-Ekbom disease Polysomnography Impulse control and compulsive behaviours Familial RLS/WED Idiopathic RLS/WED Age of onset
Introduction Restless legs syndrome/Willis–Ekbom disease (RLS/ WED) is a disorder characterized by an irresistible urge to move the legs, concomitant with an unpleasant sensation described as tingling, burning, itching, or pain in the lower limbs. Usually movement relieves these feelings, while they increase in severity at rest, in the evening or during the night [1]. In severe cases, sensory and motor symptoms can also affect the upper limbs [2]. Other characteristics that support the diagnosis include an elevated index (number per hour) of periodic limb movements during sleep (PLMS) and during wakefulness (PLMW), a positive family history of RLS/WED and a relief with dopaminergic treatment [1]. Current studies suggest that the prevalence of RLS/ WED is approximately 2–3 % in the adult Caucasian population [1, 3]. It has been found to have a high degree of familiarity, so that a family history of RLS/WED among first-degree relatives is one of the clinical features supporting the diagnosis [1]. Clinical features variability between familial and sporadic cases of RLS/WED has been compared in some of studies; an association between an early age-at-onset and the presence of a positive family
123
J Neurol
history of RLS/WED has been reported [4]. The cut-off values for differentiating ‘‘early’’ from ‘‘late’’ age of onset, however, have been rather arbitrarily defined [5, 6]. Furthermore, these studies that separated RLS/WED patients on the basis of age-at-onset did not report detailed polysomnographic data. The most frequent form of RLS/WED is called ‘‘primary’’ or ‘‘idiopathic’’ (iRLS/WED). Nevertheless, several conditions, such as anemia [7], renal failure [8] and peripheral neuropathy [9] were shown to be associated with RLS/WES, and they are considered ‘‘secondary’’ forms (sRLS/WED). Ekbom [10] first reported a high incidence of iron deficiency among patients with RLS/WED. A significant correlation between subjective ratings of RLS/ WED severity, serum ferritin and sleep efficiency was later described [11, 12]. The aim of the present retrospective study was to determine the clinical and polysomnographic characteristics of a large cohort of patients with idiopathic and secondary forms, with a focus on the distribution of age-atonset overall as well as on familiarity for RLS/WED. We also aimed to test the presence of atypical features, such as late time of symptom onset, and the risk of developing impulse control and compulsive behaviours (ICBs) in relation to dopaminergic treatment.
Methods 400 RLS/WED patients (229 women, mean age 62.05 ± 13.80 years) consecutively assessed at the Sleep Centre of the San Raffaele Hospital in Milan, Italy were enrolled for this study, from 2010 to 2014. All these patients met the criteria established by the International Restless Legs Syndrome Study Group [13]. Diagnosis was based on medical history, with a comprehensive interview for sleep disorder performed by a physician expert in sleep disorder, in addition to physical and neurological exams also carried out by a neurologist trained in sleep medicine and movement disorders. Information including age-at-onset, comorbidities, familial history, time of symptoms onset, symptoms severity, treatment and the presence of ICBs were obtained. RLS/WED patients were classified as familial if they had at least one first-degree relative who also had RLS/WED, or sporadic if they had no first-degree relatives with RLS/ WED. We considered sRLS/WED when the patient was affected by neuropathy, renal failure, amyloidosis and low ferritin values. Symptoms severity was assessed by means of the IRLS total score, while frequency of symptoms was obtained by item 7 of the IRLS [14].
123
The RLS/WED topography patterns were classified according to localization in upper limbs (UL), lower limbs (LL) or both and lateralization as symmetrical or asymmetrical limbs involvement. 122 drug naı¨ve RLS/WED patients underwent a nocturnal polysomnographic (PSG) evaluation, with an adaption night preceding the PSG recording. We considered as drug naı¨ve only the patients who were neither taking medications for RLS/WED nor other drugs that could have affected sleep, such as benzodiazepines or antidepressants. Lights-out time was based on the individual’s usual bedtime and ranged between 11.00 and 11:30 p.m. The following signals were recorded: electroencephalogram (EEG) (six channels, including F4–M1, C4–M1, 02–M1 with backup electrodes at F3, C3, O1 and M2); electrooculogram (EOG); EMG of the submentalis muscle; EMG of the right and left tibialis anterior muscles; electrocardiogram (ECG; one derivation) according to the American Academy of Sleep Medicine (AASM) scoring criteria [15]. The sleep respiratory pattern of each patient was monitored using oral and nasal airflow thermistors and/ or nasal pressure cannula, thoracic and abdominal respiratory effort strain gauge and by monitoring oxygen saturation (pulse-oximetry). Sleep stages were scored following AASM criteria [15] on 30-s epochs. Leg movements (LM) during sleep were detected following WASM criteria [16]. Patients were followed up at 3- to 6-month intervals, and eventually started a pharmacological treatment whenever symptoms conferred adverse effects on sleep, quality of life and health. Statistical analyses were performed using SPSS 13.0. ANOVA was employed to evaluate differences between iRLS/WED and sRLS/WED, and between familial and sporadic disease. When multiple comparisons were performed we employed Bonferroni correction. The Spearman’s rank correlation coefficient was calculated to test possible association between clinical, demographic and PSG data.
Results Clinical phenotype Out of 400 subjects, 229 were female (57.25 %). Duration of the disease was 18.26 ± 15.75 years, with women having a longer mean duration (male 15.62 ± 14.57 vs female 20.23 ± 16.33 years; F = 8.543, p = 0.004). RLS/ WED duration did not have any significant impact on PSG variables. On the other hand, age had a significant correlation with wake after sleep onset (WASO) (r = 0.319; p = 0.000), PLM (r = 0.361; p = 0.000) and REM
J Neurol
latency (r = 0.214; p = 0.000), and an inverse correlation with SE (r = -0.286; p = 0.002) and TST (r = -0.235; p = 0.010). Age of symptoms onset was lower in female (male 47.17 ± 16.60 vs female 43.24 ± 16.11 years; p = 0.019). It had a significant correlation with PLM (r = 0.292; p = 0.002) and REM latency (r = 0.261; p = 0.006), and an inverse correlation with sleep efficiency (SE) (r = -0.189; p = 0.041) and total sleep time (TST) (r = -0.185; p = 0.044). Nevertheless, our analysis showed that the associations found between the age of onset and polysomnographic measures of disrupted sleep are linked to aging itself more than to the age of onset. In particular, regarding SE and PLM, age had a significant effect (SE: p \ 0.005 and PLM: p \ 0.005). This is in line with the fact that RLS duration did not have any impact on PSG variables, while age was related to worse PSG characteristics. Mean scores at IRLS and ESS were 24.02 ± 6.17 and 6.34 ± 4.47, respectively, indicating that patients had a severe form of RLS/WED but without excessive daytime sleepiness. RLS severity as measured by IRLS scores correlated positively with age (r = 0.264; p \ 0.001) and with disease duration (r = 0.297; p \ 0.001). Time of onset of RLS/WED symptoms was in the evening or at bedtime in 28.04 and 37.80 % of patients, respectively. However, 21.34 % of patients reported onset of symptoms more than 1 h after falling asleep. Only 5.48 % of patients complained of symptoms onset in the afternoon, while 7.31 % reported the occurrence of symptoms both at day and at night (see Table 1). 173 patients were treated with at least one medication for RLS/WED. Of these, 89.42 % were on treatment with dopamine agonist (DA) (pramipexole, mean dosage 0.22 ± 0.16 mg; ropinirole, mean dosage 1.13 ± 0.73 mg; rotigotine, mean dosage 2 ± 0 mg), 40.38 % with clonazepam (mean dosage 0.66 ± 0.39 mg), 35.57 % with a-2d ligands (pregabalin, mean dosage 103.34 ± 68.7 mg; gabapentin, mean dosage 345.45 ± 181.86 mg) and 0.96 % with opioids. Most of the patients were on monotherapy (42.34 %), while 39.42 % used two different drugs and 18.24 % used more than two. We compared the PSG characteristics of these treated patients with the drug naı¨ve patients. The treated patients showed a lower PLM index (p \ 0.001), a longer REM latency (p \ 0.005) and a shorter REM percentage (p \ 0.05). More specifically, we found that dopamine agonist had a significant effect in reducing PLM index (p \ 0.05) and REM percentage (p \ 0.05) and in increasing REM sleep latency (p \ 0.05). ICBs were found in 13.29 % RLS/WED patients on DA therapy; most of these were related to abnormal eating (7.51 %) and sexual (4.64 %) behaviours, while gambling and punding were reported in 1.15 and 2.31 %, respectively (see Table 1).
Table 1 Clinical phenotype of RLS/WED patients Age
62.05 ± 13.80
Sex
229 women (57 %) 171 men (43 %)
Time of RLS/WED symptoms onset
Evening 28.04 % Bedtime 37.80 % More than 1 h after sleep onset 21.34 % In the afternoon 5.48 % Both at day and at night 7.31 %
Impulse control symptoms
Total 13.29 % Abnormal eating 7.51 % Sexual behaviours 4.64 % Gambling 1.15 % Punding 2.31 %
Comorbidities
Diabetes 3.93 % Psychiatric disorders 16.72 % Hypertension 13.44 % Neuropathies 4.59 % Sleep difficulties 80 % Uremia 0.32 %
Localization of symptoms
Bilateral and symmetric lower limbs (LL) 74.7 % Asymmetric LL 4.1 % Upper limbs 0.7 % Four limbs symmetrical 17.8 % Asymmetrical 2.7 %
Age-at-onset
44.93 ± 16.42
Considering comorbidities, most of the patients complained of some kind of sleep difficulties (80 %); other frequent diseases were psychiatric disorders (16.72 %), hypertension (13.44 %), neuropathies (4.59 %) and diabetes (3.93 %). Only 0.32 % of patients reported the presence of uremia. In 146 untreated patients, information on the localization of symptoms were collected. Bilateral and symmetric lower limb (LL) location was the most frequent (74.7 %), while 4.1 % of participants exhibited asymmetric LL localization. Only one patient (0.7 %) showed symptoms confined to the upper limbs (UL), symmetrically. Four limbs involvement was found in 20.5 % of patients, and was symmetrical in 86.8 % of cases. Overall, 6.9 % of patients complained of asymmetrical symptoms. The severity of RLS/WED as measured with IRLS was similar across the patients, independent of the localization of symptoms. Conversely, IRLS values correlated positively with PLM (r = 0.205; p = 0.044), N2 (r = 0.219; p = 0.030) and negatively with REM sleep (r = -0.243; p = 0.15). The single regression analyses showed that PLM index (p \ 0.05), duration of disease (p \ 0.001) and ESS values (p \ 0.05) are predictors of IRLS values.
123
J Neurol
We assessed the presence of pregnancy in the personal history of the female patients, and its effects on RLS/WED symptoms. 34.4 % had no pregnancy in their life, while 31.8 % reported no effect, 4.6 % mild worsening, 7.2 % moderate worsening, 6.7 % severe worsening and 15.4 % had no recall of this effect. Familial vs. sporadic RLS/WED Familial RLS/WED was present in 186 (53.5 %) patients, while 214 (46.5 %) had a sporadic form. Mean age-at-onset differed as a function of presence/absence of a familial history of RLS/WED (40.44 ± 16.43 vs. 49.03 ± 15.33 years; p \ 0.001). Familial RLS/WED patients were also younger (60.47 ± 14.54 vs. 63.41 ± 12.96 years; p \ 0.05). These differences were significant even when adjusted for age and gender. Polysomnographic characteristics were similar in both groups [including sleep latency (SL); WASO; SE; TST; number of awakenings: PLM index; stage N1; stage N2; stage N3; REM; REM latency]. Likewise, RLS/WED severity as reported by the IRLS scale was similar between the two groups. The two groups did not differ even in the ESS values and in the serum ferritin values (see Table 2). Table 2 Familial vs. sporadic RLS/WED Familial
Idiopathic vs. secondary RLS/WED 277 out of 400 subjects had iRLS/WED (49.50 % female). No differences were found in the age-at-onset between idiopathic and secondary forms (44.89 ± 16.35 vs. 45.02 ± 16.64 years) or in the age of the patients (62.10 ± 13.83 vs. 61.93 ± 13.74 years). Regarding the PSG findings, PLM index was significantly higher in iRLS/WED (50.07 ± 47.72 vs. 32.39 ± 41.46; p = 0.047), while REM duration was longer in iRLS/WED (20.08 ± 6.22 vs 17.48 ± 6.91 min; p = 0.043). These differences were not significant when adjusted for age and gender. The other PSG parameters did not differ between the two groups. IRLS and ESS values were higher in sRLS/WED (IRLS score: 25.365 ± 6.35 vs 23.33 ± 5.98, p \ 0.05; ESS score 7.43 ± 5.22 vs 5.79 ± 3.93, p \ 0.05), even when adjusted for age and gender, while ferritin values were similar between the two groups (see Table 3).
Discussion The aim of this study was to describe a large cohort of RLS/WED patients, comparing hereditary and sporadic as well as idiopathic and secondary disease. We focused on Table 3 Idiopathic vs. secondary RLS/WED
Sporadic
p
Idiopathic
SL
28.13 ± 37.22
28.13 ± 25.46
NS
SL
WASO
91.57 ± 85.30
114.95 ± 82.63
NS
WASO
SE TST
73.39 ± 17.91
72.67 ± 17.43
NS
SE
360.15 ± 96.84
369.22 ± 92.28
NS
TST
Secondary
p
28.64 ± 38.47
26.84 ± 53.03
NS
102.61 ± 84.06
98.28 ± 86.64
NS
72.45 ± 17.80
74.22 ± 17.55
NS
362.39 ± 94.73
366.34 ± 95.80
NS
AWK
14.29 ± 7.80
14.00 ± 6.53
NS
AWK
14.24 ± 6.61
14.07 ± 8.42
NS
PLMSI
45.30 ± 49.94
40.59 ± 40.07
NS
PLMSI
50.07 ± 47.72
32.39 ± 41.46
NS
Stage N1
5.58 ± 4.08
6.64 ± 6.05
NS
Stage N1
6.19 ± 5.36
5.71 ± 4.29
NS
Stage N2
52.31 ± 12.59
49.58 ± 13.84
NS
Stage N2
50.06 ± 13.46
53.23 ± 12.42
NS
Stage N3
22.87 ± 9.91
25.20 ± 11.61
NS
Stage N3
23.71 ± 11.27
24.03 ± 9.54
NS
Stage R
19.61 ± 5.87
18.49 ± 7.48
NS
Stage R
20.08 ± 6.22
17.48 ± 6.91
NS
130.88 ± 83.97 96.90 ± 119.34
129.93 ± 69.56 115.12 ± 126.66
NS NS
Stage R latency Ferritin
131.37 ± 83.75 160.69 ± 139.82
128.91 ± 67.56 49.30 ± 65.89
NS NS
IRLS
24.36 ± 6.39
23.68 ± 5.95
NS
IRLS
23.33 ± 5.98
25.365 ± 6.35
\0.05§*
ESS
6.16 ± 4.35
6.51 ± 4.58
NS
\0.05§*
Stage R latency Ferritin
Age Age of onset
60.47 ± 14.54 40.44 ± 16.43
63.41 ± 12.96 49.02 ± 15.33
ESS
5.79 ± 3.93
7.43 ± 5.22
§
Age
62.10 ± 13.83
61.93 ± 13.74
NS
§
Age of onset
44.89 ± 16.35
45.02 ± 16.64
NS
\0.05 * \0.05 *
SL sleep latency, WASO wake after sleep onset, SE percent sleep efficiency, TST total sleep time, AWK awakenings, PLMSI periodic limb movements of sleep index, stage N1 stage non-REM 1, stage N2 stage non-REM 2, stage N3 stage non-REM3, stage R stage REM, IRLS international restless legs syndrome rating scale, ESS Epworth Sleepiness Scale § Adjusted for age and gender
SL sleep latency, WASO wake after sleep onset, SE percent sleep efficiency, TST total sleep time, AWK awakenings, PLMSI periodic limb movements of sleep index, stage N1 stage non-REM 1, stage N2 stage non-REM 2, stage N3 stage non-REM3, stage R stage REM, IRLS international restless legs syndrome rating scale, ESS Epworth Sleepiness Scale § Adjusted for age and gender
* Adjusted for Bonferroni correction
* Adjusted for Bonferroni correction
123
J Neurol
the age of onset, the polysomnographic findings and the clinical signs and symptoms of RLS/WED. Our study revealed the presence of several atypical features in a large proportion of patients with RLS/WED, which may pose a diagnostic challenge. Firstly, even though most of the patients reported the onset of symptoms in the evening or at bedtime, a relevant percentage complained of symptom beginning 1 h or more after sleep onset. This is of peculiar importance when investigating the presence of symptoms suggestive of RLS/WED in patients with sleep maintenance insomnia. Secondly, we found a higher percentage of ICBs than reported in most of the previous works (13.29 vs. 10 % in Lipford et al. [35], 8.6 % in Ondo et al. [36], 12.4 % in Pourcher et al. [37], 17 % in Cornelius et al. [38]). This is of relevance considering that, in our cohort of patients, the mean dosage of pramipexole was much lower (0.22 mg, compared to 1.33 mg in the study by Lipford et al.) [35], and sexual behaviours were more frequent than previously described [37, 39, 40]. Our results point to the importance of taking a careful history of risk factors when offering a treatment with DA, even if at low dosage, forewarning the patients of potential side effects and planning a close follow-up. The use of a a-2-d calcium-channel ligand should also be considered [41]. The importance of a careful evaluation is furthermore underlined by the frequent occurrence of sleep complaints, experienced by over 80 % of the patients. In these cases, in fact, RLS/WED symptoms may go unnoticed because of more manifest sleep disorders. Moreover, the age-at-onset distribution in the whole sample of our study did not show the diphasic distribution described in previous works. Defining a cut-off age for distinguishing early onset from late onset RLS/WED has always been intricate, and previous works posed a mark at different ages, varying from 36 years [29] to 45 years [20]. One possible explanation for these differences is that the age for differentiating ‘‘early’’ from ‘‘late’’ had been distorted because of smaller samples. We, therefore, suggest that physicians do not rely solely on age of onset for identifying the etiology of RLS/WED. Nevertheless, a younger onset of RLS/WED symptoms should raise suspicions for a familial form. On the other side, some of our findings are in line with the literature data, and can therefore add information that can support hypotheses about the pathophysiology of RLS/ WED. The severity of RLS/WED as measured with IRLS correlated positively with age and with disease duration, confirming the view of a slow progression of symptoms [20]. Furthermore, notwithstanding the presence of high IRLS scores, mean ESS scores were definitely below the
threshold for a significant daytime sleepiness [1]. This finding supports the hypothesis that RLS/WED enhances alertness despite sleep loss, thanks to hyperarousal [1, 21]. According to this theoretical view, hyperarousal is caused by an increased level of glutamatergic activity in the thalamus. This would lead, on one side, to an increased waking time during sleep, and on the other side, to a decreased daytime somnolence. Hyperarousal can also explain the persistence of insomnia in patients with RLS/ WED otherwise successfully treated with dopamine agonists [22]. Previous surveys differed somewhat regarding gender distribution, with some reporting a higher prevalence for women [23, 24] while others did not [25, 26]. In our sample, 57.25 % of patients were females, supporting the view that female sex is a risk factor for RLS/WED [1]. Female patients also showed a younger age of onset. Talking about age of onset of the disease, our study shows that patients with a familial history of RLS/WED were almost 10 years younger than those without, confirming previous findings [5, 20, 29, 30], and supporting the hypothesis that early onset RLS/WED may have a genetic component. Besides that, we did not find any significant difference in clinical or PSG characteristics between familial and sporadic RLS/WED. We also found a lower percentage of familial RLS/WED than previously reported (53.5 vs up to 60–65 % of cases [6, 27]). These differences between studies may be ascribed to different definitions of ‘‘familial’’. In fact, relatives are not always available for a direct assessment by sleep specialist, and even less for a complete PSG evaluation, so the definition of ‘‘familiarity’’ relies on the physician’s feeling about the patients’ story about their relatives’ symptoms. Since the genetic of RLS/ WED is still not completely clarified [28], we considered a positive family history only if first a grade relative was diagnosed affected by the disease. On the contrary, the age of onset of secondary and idiopathic RLS/WED was similar, although previous works found an earlier onset of symptoms in the latter [29]. Even in this case, an explanation for this discrepancy may lie in the different definition of ‘‘secondary’’ RLS/WED, encompassing association with uremia, rheumatoid arthritis, amyloidosis, iron deficiency, folate and B12 deficiency [11, 31–34]. Regarding our sample, there was a particularly low amount of cases of RLS/WED due to ‘‘end-stage renal disease’’, with a relatively high number of neuropathy-related RLS/WED. This could be explained by the older age of sRLS/WED patients of our sample, or due to a selection bias. The comorbidities shown by our sample were otherwise similar to what were expected from the literature data [17–19] (see Table 1). Regarding the polysomnographic findings, a lower PLM index was found in treated patients, confirming the efficacy
123
J Neurol
of treatment. They also displayed a longer REM latency and a shorter REM stage duration. This may be mainly due to the usage of DA, that can delay REM sleep latency and decrease total percentage of REM sleep [42, 43]. Our study shows some biases. First, being retrospective and not population based, it may be due to selection bias. The sample included either patients with and without DA treatment or other medications that could have influenced in some part of the results. Furthermore, since a significant amount of patients was under a polytherapy we did not perform sub-group analyses to avoid biases due to skewed data in small subgroups of patients. The duration of treatment also was very different among those patients. Moreover, we were not able to interview relatives of patients, and this may have caused erroneous placement into the familial or sporadic groups. On the other hand, this work has some strengths. It displays a large cohort of patients with RLS/WED who were evaluated by the same team of physicians in a relatively short period, with a substantial amount undergoing PSG evaluation, both with and without treatment. The presence of atypical symptoms in a large proportion of patients is of peculiar importance and points to the importance of obtaining a thorough history.
Conclusions RLS/WED is an easy recognizable disease when all diagnostic criteria are met, but a large proportion of patients can have atypical features, therefore making the diagnosis challenging. Some cases can be missed even when the patient refers to a sleep specialist, as revealed by the partial absence of daytime symptoms, the high comorbidity with insomnia and other sleep complaints and the high percentage of symptoms beginning after sleep onset. The dichotomy between early and late onset disease seems to be less sharp than previously reported. This draws attention on the importance of a careful evaluation of the patient, to recognize potentially treatable secondary forms of RLS/WED. The overall similarity of clinical characteristics and the course of the disease in idiopathic, hereditary and sporadic RLS/WED could suggest a common pathological pathway. The risk of developing ICBs seems to occur more frequently and with lower dosages of dopaminergic drugs than previously reported. The physician should be aware of this risk and consider the use of alternative medications. Compliance with ethical standards Conflicts of interest of interest.
123
The authors declare that they have no conflict
Ethical standards All participants provided written informed consent to the experimental procedure, which was previously approved by the local ethical committee.
References 1. Allen RP, Picchietti DL, Garcia-Borreguero D, Ondo WG, Walters AS, Winkelman JW, Zucconi M, Ferri R, Trenkwalder C, Lee HB, International Restless Legs Syndrome Study Group (2014) Restless legs syndrome/Willis-Ekbom disease diagnostic criteria: updated International Restless Legs Syndrome Study Group (IRLSSG) consensus criteria—history, rationale, description, and significance. Sleep Med 15(8):860–873 2. Chabli A, Michaud M, Montplaisir J (2000) Periodic arm movements in patients with the restless legs syndrome. Eur Neurol 44(3):133–138 3. Garcia-Borreguero D, Egatz R, Winkelmann J, Berger K (2006) Epidemiology of restless legs syndrome: the current status. Sleep Med Rev 10(3):153–167 4. Winkelmann J (2002) The genetics of restless legs syndrome. Sleep Med 3(Suppl):S9–S12 5. Winkelmann J, Muller-Myhsok B, Wittchen HU, Hock B, Prager M, Pfister H, Strohle A, Eisensehr I, Dichgans M, Gasser T, Trenkwalder C (2002) Complex segregation analysis of restless legs syndrome provides evidence for an autosomal dominant mode of inheritance in early age at onset families. Ann Neurol 52(3):297–302 6. Allen RP, La Buda MC, Becker P, Earley CJ (2002) Family history study of the restless legs syndrome. Sleep Med 3(Suppl):S3–S7 7. Aul EA, Davis BD, Rodnitzky RL (1998) The importance of formal serum iron studies in the assessment of restless legs syndrome. Neurology 51(3):912 8. Winkelman JW, Chertow GM, Lazarus JM (1996) Restless legs syndrome in end-stage renal disease. Am J Kidney Dis 28(3):372–378 9. Ondo W, Jankovic J (1996) Restless legs syndrome: clinicoetiologic correlates. Neurology 47:1435–1441 10. Ekbom KA (1960) Restless legs syndrome. Neurology 10:868–873 11. O’Keeffe ST, Gavin K, Lavan JN (1994) Iron status and restless legs syndrome in the elderly. Age Ageing 23(3):200–203 12. Sun ER, Chen CA, Ho G, Earley CJ, Allen RP (1998) Iron and the restless legs syndrome. Sleep 21(4):371–377 13. Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS, Montplaisi J, Restless Legs Syndrome Diagnosis and Epidemiology workshop at the National Institutes of Health, International Restless Legs Syndrome Study Group (2003) 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 4(2):101–119 14. Walters AS, LeBrocq C, Dhar A, Hening W, Rosen R, Allen RP, Trenkwalder C, International Restless Legs Syndrome Study Group (2003) Validation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med 4(2):121–132 15. Iber C, Ancoli-Israel S, Chesson AL, Quan SF, For the American Academy of Sleep Medicine (2007) The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications, 1st edn. American Academy of Sleep Medicine, Westchester 16. Zucconi M, Ferri R, Allen R, Baier PC, Bruni O, Chokroverty S, Ferini-Strambi L, Fulda S, Garcia-Borreguero D, Hening WA,
J Neurol
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Hirshkowitz M, Ho¨gl B, Hornyak M, King M, Montagna P, Parrino L, Plazzi G, Terzano MG, International Restless Legs Syndrome Study Group (IRLSSG) (2006) 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 7(2):175–183 Benediktsdottir B, Janson C, Lindberg E, Arnardo´ttir ES, Olafsson I, Cook E, Thorarinsdottir EH, Gislason T (2010) Prevalence of restless legs syndrome among adults in Iceland and Sweden: Lung function, comorbidity, ferritin, biomarkers and quality of life. Sleep Med 11(10):1043–1048 Szentkira´lyi A, Vo¨lzke H, Hoffmann W, Trenkwalder C, Berger K (2014) Multimorbidity and the risk of restless legs syndrome in 2 prospective cohort studies. Neurology 82(22):2026–2033 Ulfberg J, Bjorvatn B, Leissner L, Gyring J, Karlsborg M, Regeur L, Skeidsvoll H, Polo O, Partinen M, Nordic RLS Study Group (2007) Comorbidity in restless legs syndrome among a sample of Swedish adults. Sleep Med 8(7–8):768–772 Allen RP, Earley CJ (2000) Defining the phenotype of the restless legs syndrome (RLS) using age-of-symptom-onset. Sleep Med 1(1):11–19 Riemann D, Spiegelhalder K, Feige B, Voderholzer U, Berger M, Perlis M, Nissen C (2010) The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Med Rev 14(1):19–31 Ferri R, Cosentino FI, Manconi M, Rundo F, Bruni O, Zucconi M (2014) Increased electroencephalographic high frequencies during the sleep onset period in patients with restless legs syndrome. Sleep 37(8):1375–1381 Phillips B, Hening W, Britz P, Mannino D (2006) Prevalence and correlates of restless legs syndrome: results from the 2005 National Sleep Foundation Poll. Chest 129(1):76–80 Rothdach AJ, Trenkwalder C, Haberstock J, Keil U, Berger K (2000) Prevalence and risk factors of RLS in an elderly population: the MEMO study. Memory and morbidity in Augsburg elderly. Neurology 54(5):1064–1068 Lavigne GJ, Montplaisir JY (1994) Restless legs syndrome and sleep bruxism: prevalence and association among Canadians. Sleep 17(8):739–743 Phillips B, Young T, Finn L, Asher K, Hening WA, Purvis C (2000) Epidemiology of restless legs symptoms in adults. Arch Internal Med 160(14):2137–2141 Bonati M, Ferini-Strambi L, Aridon P, Oldani A, Zucconi M, Casari G (2003) Autosomal dominant restless legs syndrome maps an chromosome 14q. Brain 126(Pt 6):1485–1492 Schormair B, Winkelmann J (2011) Genetics of restless legs syndrome: Mendelian, complex, and everything in between. Sleep Med Clin 6(2):203–215 Whittom S, Dauvilliers Y, Pennestri MH, Vercauteren F, Molinari N, Petit D, Montplaisir J (2007) Age-at-onset in restless legs
30.
31.
32.
33. 34. 35.
36.
37.
38.
39.
40.
41.
42.
43.
syndrome: a clinical and polysomnographic study. Sleep Med 9(1):54–59 Hanson M, Honour M, Singleton A, Crawley A, Singleton A, Hardy J, Gwinn-Hardy K (2004) Analysis of familial and sporadic restless legs syndrome in age of onset, gender, and severity features. J Neurol 251(11):1398–1401 Collado-Seidel V, Kohnen R, Samtleben W, Hillebrand GF, Oertel WH, Trenkwalder C (1998) Clinical and biochemical findings in uremic patients with and without restless legs syndrome. Am J Kidney Dis 31(2):324–328 Reynolds G, Blake DR, Pall HS, Williams A (1986) Restless leg syndrome and rheumatoid arthritis. Br Med J (Clin Res Ed) 292(6521):659–660 Heinze EG, Frame B, Fine G (1967) Restless legs and orthostatic hypotension in primary amyloidosis. Arch Neurol 16(5):497–500 Botez MI, Lambert B (1977) Folate-deficiency and restless-legssyndrome in pregnancy. N Engl J Med 297(12):670 Lipford MC, Silber MH (2012) Long-term use of pramipexole in the management of restless legs syndrome. Sleep Med 13(10):1280–1285 Ondo WG, Lai D (2008) Predictors of impulsivity and reward seeking behavior with dopamine agonists. Parkinsonism Relat Disord 14(1):28–32 Pourcher E, Re´millard S, Cohen H (2010) Compulsive habits in restless legs syndrome patients under dopaminergic treatment. J Neurol Sci 290(1–2):52–56 Cornelius JR, Tippmann-Peikert M, Slocumb NL, Frerichs CF, Silber MH (2010) Impulse control disorders with the use of dopaminergic agents in restless legs syndrome: a case-control study. Sleep 33(1):81–87 Schreglmann SR, Gantenbein AR, Eisele G, Baumann CR (2012) Transdermal rotigotine causes impulse control disorders in patients with restless legs syndrome. Parkinsonism Relat Disord 18(2):207–209 Voon V, Schoerling A, Wenzel S, Ekanayake V, Reiff J, Trenkwalder C, Sixel-Do¨ring F (2011) Frequency of impulse control behaviours associated with dopaminergic therapy in restless legs syndrome. BMC Neurol 11:117 Garcia-Borreguero D, Kohnen R, Silber MH, Winkelman JW, Earley CJ, Ho¨gl B, Manconi M, Montplaisir J, Inoue Y, Allen RP (2013) The long-term treatment of restless legs syndrome/WillisEkbom disease: evidence-based guidelines and clinical consensus best practice guidance: a report from the International Restless Legs Syndrome Study Group. Sleep Med 14(7):675–684 Montplaisir J, Nicolas A, Denesle R et al (1999) Restless legs syndrome improved by pramipexole. A double-blind randomized trial. Neurology 52:938–943 Saletu M, Anderer P, Saletu-Zyhlarz G et al (2002) Acute placebo-controlled sleep laboratory studies and clinical follow-up with pramipexole in restless legs syndrome. Eur Arch Psychiatry Clin Neurosci 252:185–194
123
