European Journal of Neurology 2014
doi:10.1111/ene.12462
Association between restless legs syndrome and migraine: a population-based study Stefano Zanignia, Giulia Gianninia, Roberto Melottia, Cristian Pattaroa, Federica Provinib,c, Sabina Cevolib, Maurizio F. Facherisa, Pietro Cortellib,c and Peter P. Pramstallera a
Center for Biomedicine, European Academy Bozen/Bolzano (EURAC) (Affiliated Institute of the University of L€ ubeck), Bozen/Bolzano; IRCCS Institute of Neurological Sciences of Bologna, Bologna; and cDepartment of Biomedical and NeuroMotor Sciences (DiBiNeM),
b
Alma Mater Studiorum – University of Bologna, Bologna, Italy
Keywords:
comorbidity, headache, migraine, restless legs syndrome, sleep disorders Received 19 January 2014 Accepted 7 April 2014
Background and purpose: A higher prevalence of restless legs syndrome (RLS) in migraineurs has been reported in clinical samples and in two large-scale clinical trials performed on healthcare workers but general population-based studies on this topic are lacking. The aim of this study was to assess the association between migraine and RLS in an Italian rural adult population-based setting. Methods: The presence of migraine and RLS was assessed via a computer-assisted personal interview and self-administered questionnaires according to current diagnostic criteria in 1567 participants of a preliminary phase of an adult populationbased study performed in South Tyrol, Italy. Results: Migraineurs had an increased risk of having RLS also after adjustment for confounding factors such as age, sex, major depression, anxiety and sleep quality (odds ratio 1.79; confidence interval 1.00–3.19; P = 0.049). This association was not modified by aura status and possible causes of secondary RLS. RLS was not significantly associated with tension-type headache. Conclusions: Restless legs syndrome and migraine were associated in our rural adult population. This association could be explained by a possible shared pathogenic pathway which would implicate new management strategies of these two disorders.
Background Restless legs syndrome (RLS) is a sensory-motor disorder affecting 4%–29% of the general population [1], characterized by an uncomfortable sensation usually in the legs; it typically worsens at evening and ameliorates with movement. RLS is classified amongst sleeprelated movement disorders in the International Classification of Sleep Disorders ICSD-II [2] and current diagnostic criteria were established in 2003 by an international task force (International Restless Legs Syndrome Study Group – IRLSSG) [3]. RLS pathophysiology is unclear and peripheral and central mechanisms within the nervous system have been hypothesized. Because of the effect of dopaminergic therapy, which is shown to improve RLS symptomatology, and of functional neuroimaging studies, the involvement of the dopaminergic pathway in RLS Correspondence: S. Zanigni, Center for Biomedicine, EURAC Research, Via Galvani 31, I-39100 Bolzano, Italy (tel.: +39 0471 055 500; fax: +39 0471 055 599; e-mail: [email protected]).
© 2014 The Author(s) European Journal of Neurology © 2014 EFNS
pathogenesis has been suggested [4,5]. Migraine is a primary headache [6] affecting about 15% of the general population [7]. Dopamine has also been involved in migraine pathogenesis [8,9]. Migraine has been associated with sleep disorders [10]; in particular clinical studies and two large-scale population-based studies performed on healthcare professionals have evidenced a higher prevalence of RLS in migraineurs compared with controls [11–19] (Table 1). It has also been reported that RLS correlates positively with migraine frequency and with the burden for patients measured by the Migraine Disability Assessment Scale (MIDAS) [16] (Table 1). A lack of association between RLS and migraine with aura (MwA), tension-type headache (TTH) and cluster headache (CH) has been supposed [14,20,21]. The aim of this study was to assess the association between migraine and RLS in a rural adult population without restriction for age, sex, profession and comorbidity. A secondary outcome was to evaluate the association between RLS and TTH and aura status to corroborate the specificity of this association.
1
2
S. Zanigni et al.
Table 1 Studies on the association between migraine and RLS Author (year)
Study design
Study sample
Sample size
Results
Young et al. (2003) [11] Rhode et al. (2007) [12]
Cross-sectional Case control
Clinical Clinical
D’Onofrio et al. (2008) [13]
Case control
Clinical
RLS 34% in migraineurs RLS 17.3% migraineurs vs. 5.6% controls (P < 0.001) RLS 22.4% headache patients vs. 8.3% controls (P < 0.002)
Chen et al. (2010) [14]
Cross-sectional
Clinical
Suzuki et al. (2011) [15]
Cross-sectional
Clinical
Lucchesi et al. (2012) [16]
Cross-sectional
Clinical
Seidel et al. (2012) [19]
Cross-sectional
Sch€ urks et al. (2012) [18]
Cross-sectional and cohort longitudinal Cross-sectional
Clinical (children and adolescents) Population-based (women ≥45 years old) Population-based (pooled analysis from two different samples with men 40–84 years old and ≥55 years old)
50 migraineurs 411migraineurs 411 controls 200 headache patients 120 controls 1041 headache patients (772 migraineurs) 262 migraineurs 163 controls 277 migraineurs 200 controls 111 migraineurs 2 control groups (n = 73 and 108 respectively) 31 370
Winter et al. (2013) [17]
22 926
RLS 11.4% migraineurs, significantly higher than TTH and CH (P = 0.002) RLS 13.7% migraineurs vs. 1.8% controls (P < 0.0001) RLS 22.7% migraineurs vs. 7.5% controls (P < 0.0001) RLS in 22% migraineurs vs. 5% in control group 1 and 8% in group 2 (P < 0.001) OR 1.22 for ‘any migraine’ (95% CI 1.13–1.32; P < 0.0001) and 1.35 for ‘new migraine’ (95% CI 1.10–1.54; P < 0.0001) OR 1.20 (CI 95% 1.04–1.38; P value not reported)
RLS, restless legs syndrome; TTH, tension-type headache; CH, cluster headache; OR, odds ratio; CI, confidence interval.
Materials and methods The current study is a cross-sectional analysis carried out within a preliminary phase of an ongoing population-based study targeting the whole adult population (≥18 years old) in Val Venosta, an Italian and German speaking valley located in northern Italy. The overall aims of the study were to investigate the role of the interaction between genetic and environmental factors in cardiovascular, neurological and metabolic conditions. The study was approved by the local ethics committee (Comprensorio Sanitario di Bolzano). All the eligible population was invited in the study center to a face-toface clinical assessment performed by a trained studynurse and all participants gave written informed consent. The first 1567 participants were enrolled from August 2011 until October 2012 in a study center located in the valley, with a participation rate that reached about 35%. The comparison between our sample and the adult population living in the valley showed no statistically significant differences in sex and age ranges.
Headache assessment
Headache was assessed through computer-assisted interviewer-administered questionnaire based on the
International Classification of Headache Disorders 2nd edition (ICHD-II) [6] criteria. To discriminate between headache sufferers and non-sufferers, participants were asked to answer the question ‘Did you suffer from headache in the last 12 months?’ In the case of a positive answer, participants filled a detailed questionnaire on headache frequency, attack duration, pain quality, severity, localization and associated features (i.e. visual-sensory-motor-aphasic aura, nausea, vomiting, osmo-phono-photophobia, local autonomic symptoms, influence of routine physical activity on pain and burden on daily activities). Migraine and TTH were diagnosed by using a computer algorithm based on ICHD-II criteria [6] for migraine, migraine without aura (MwoA), MwA and TTH. RLS assessment
Restless legs syndrome was assessed by a self-administered questionnaire containing four questions based on the Italian and German translation of the four essential IRLSSG criteria [3]. RLS sufferers were considered to be subjects who answered positively to all four questions, referring to the last month: (i) Do you have unpleasant sensations such as crawling or pain in your legs combined with an urge or need to move © 2014 The Author(s) European Journal of Neurology © 2014 EFNS
Association between RLS and migraine
your legs? (ii) Do these feelings/symptoms occur mainly or only at rest? (iii) Does movement improve these unpleasant sensations? (iv) Are these symptoms worse in the evening or at night than in the morning? RLS sufferers fulfilled the IRLSSG rating scale [22]. Covariate assessment
Depression was assessed by means of the Center for Epidemiologic Studies Depression (CES-D) scale [23] with a pathological score ≥23 suggesting the presence of major depression [24]. Sleep quality was assessed by the Pittsburgh Sleep Quality Index (PSQI) [25] considering as ‘poor sleep quality’ total scores of >5. Anxiety was assessed by the State-Trait Anxiety Inventory (STAI) scale [26]. CES-D, PSQI and STAI scales were self-administered. Possible secondary causes of RLS were also considered [3] evaluating the presence of low blood ferritin levels (ferritin
doi:10.1111/ene.12462
Association between restless legs syndrome and migraine: a population-based study Stefano Zanignia, Giulia Gianninia, Roberto Melottia, Cristian Pattaroa, Federica Provinib,c, Sabina Cevolib, Maurizio F. Facherisa, Pietro Cortellib,c and Peter P. Pramstallera a
Center for Biomedicine, European Academy Bozen/Bolzano (EURAC) (Affiliated Institute of the University of L€ ubeck), Bozen/Bolzano; IRCCS Institute of Neurological Sciences of Bologna, Bologna; and cDepartment of Biomedical and NeuroMotor Sciences (DiBiNeM),
b
Alma Mater Studiorum – University of Bologna, Bologna, Italy
Keywords:
comorbidity, headache, migraine, restless legs syndrome, sleep disorders Received 19 January 2014 Accepted 7 April 2014
Background and purpose: A higher prevalence of restless legs syndrome (RLS) in migraineurs has been reported in clinical samples and in two large-scale clinical trials performed on healthcare workers but general population-based studies on this topic are lacking. The aim of this study was to assess the association between migraine and RLS in an Italian rural adult population-based setting. Methods: The presence of migraine and RLS was assessed via a computer-assisted personal interview and self-administered questionnaires according to current diagnostic criteria in 1567 participants of a preliminary phase of an adult populationbased study performed in South Tyrol, Italy. Results: Migraineurs had an increased risk of having RLS also after adjustment for confounding factors such as age, sex, major depression, anxiety and sleep quality (odds ratio 1.79; confidence interval 1.00–3.19; P = 0.049). This association was not modified by aura status and possible causes of secondary RLS. RLS was not significantly associated with tension-type headache. Conclusions: Restless legs syndrome and migraine were associated in our rural adult population. This association could be explained by a possible shared pathogenic pathway which would implicate new management strategies of these two disorders.
Background Restless legs syndrome (RLS) is a sensory-motor disorder affecting 4%–29% of the general population [1], characterized by an uncomfortable sensation usually in the legs; it typically worsens at evening and ameliorates with movement. RLS is classified amongst sleeprelated movement disorders in the International Classification of Sleep Disorders ICSD-II [2] and current diagnostic criteria were established in 2003 by an international task force (International Restless Legs Syndrome Study Group – IRLSSG) [3]. RLS pathophysiology is unclear and peripheral and central mechanisms within the nervous system have been hypothesized. Because of the effect of dopaminergic therapy, which is shown to improve RLS symptomatology, and of functional neuroimaging studies, the involvement of the dopaminergic pathway in RLS Correspondence: S. Zanigni, Center for Biomedicine, EURAC Research, Via Galvani 31, I-39100 Bolzano, Italy (tel.: +39 0471 055 500; fax: +39 0471 055 599; e-mail: [email protected]).
© 2014 The Author(s) European Journal of Neurology © 2014 EFNS
pathogenesis has been suggested [4,5]. Migraine is a primary headache [6] affecting about 15% of the general population [7]. Dopamine has also been involved in migraine pathogenesis [8,9]. Migraine has been associated with sleep disorders [10]; in particular clinical studies and two large-scale population-based studies performed on healthcare professionals have evidenced a higher prevalence of RLS in migraineurs compared with controls [11–19] (Table 1). It has also been reported that RLS correlates positively with migraine frequency and with the burden for patients measured by the Migraine Disability Assessment Scale (MIDAS) [16] (Table 1). A lack of association between RLS and migraine with aura (MwA), tension-type headache (TTH) and cluster headache (CH) has been supposed [14,20,21]. The aim of this study was to assess the association between migraine and RLS in a rural adult population without restriction for age, sex, profession and comorbidity. A secondary outcome was to evaluate the association between RLS and TTH and aura status to corroborate the specificity of this association.
1
2
S. Zanigni et al.
Table 1 Studies on the association between migraine and RLS Author (year)
Study design
Study sample
Sample size
Results
Young et al. (2003) [11] Rhode et al. (2007) [12]
Cross-sectional Case control
Clinical Clinical
D’Onofrio et al. (2008) [13]
Case control
Clinical
RLS 34% in migraineurs RLS 17.3% migraineurs vs. 5.6% controls (P < 0.001) RLS 22.4% headache patients vs. 8.3% controls (P < 0.002)
Chen et al. (2010) [14]
Cross-sectional
Clinical
Suzuki et al. (2011) [15]
Cross-sectional
Clinical
Lucchesi et al. (2012) [16]
Cross-sectional
Clinical
Seidel et al. (2012) [19]
Cross-sectional
Sch€ urks et al. (2012) [18]
Cross-sectional and cohort longitudinal Cross-sectional
Clinical (children and adolescents) Population-based (women ≥45 years old) Population-based (pooled analysis from two different samples with men 40–84 years old and ≥55 years old)
50 migraineurs 411migraineurs 411 controls 200 headache patients 120 controls 1041 headache patients (772 migraineurs) 262 migraineurs 163 controls 277 migraineurs 200 controls 111 migraineurs 2 control groups (n = 73 and 108 respectively) 31 370
Winter et al. (2013) [17]
22 926
RLS 11.4% migraineurs, significantly higher than TTH and CH (P = 0.002) RLS 13.7% migraineurs vs. 1.8% controls (P < 0.0001) RLS 22.7% migraineurs vs. 7.5% controls (P < 0.0001) RLS in 22% migraineurs vs. 5% in control group 1 and 8% in group 2 (P < 0.001) OR 1.22 for ‘any migraine’ (95% CI 1.13–1.32; P < 0.0001) and 1.35 for ‘new migraine’ (95% CI 1.10–1.54; P < 0.0001) OR 1.20 (CI 95% 1.04–1.38; P value not reported)
RLS, restless legs syndrome; TTH, tension-type headache; CH, cluster headache; OR, odds ratio; CI, confidence interval.
Materials and methods The current study is a cross-sectional analysis carried out within a preliminary phase of an ongoing population-based study targeting the whole adult population (≥18 years old) in Val Venosta, an Italian and German speaking valley located in northern Italy. The overall aims of the study were to investigate the role of the interaction between genetic and environmental factors in cardiovascular, neurological and metabolic conditions. The study was approved by the local ethics committee (Comprensorio Sanitario di Bolzano). All the eligible population was invited in the study center to a face-toface clinical assessment performed by a trained studynurse and all participants gave written informed consent. The first 1567 participants were enrolled from August 2011 until October 2012 in a study center located in the valley, with a participation rate that reached about 35%. The comparison between our sample and the adult population living in the valley showed no statistically significant differences in sex and age ranges.
Headache assessment
Headache was assessed through computer-assisted interviewer-administered questionnaire based on the
International Classification of Headache Disorders 2nd edition (ICHD-II) [6] criteria. To discriminate between headache sufferers and non-sufferers, participants were asked to answer the question ‘Did you suffer from headache in the last 12 months?’ In the case of a positive answer, participants filled a detailed questionnaire on headache frequency, attack duration, pain quality, severity, localization and associated features (i.e. visual-sensory-motor-aphasic aura, nausea, vomiting, osmo-phono-photophobia, local autonomic symptoms, influence of routine physical activity on pain and burden on daily activities). Migraine and TTH were diagnosed by using a computer algorithm based on ICHD-II criteria [6] for migraine, migraine without aura (MwoA), MwA and TTH. RLS assessment
Restless legs syndrome was assessed by a self-administered questionnaire containing four questions based on the Italian and German translation of the four essential IRLSSG criteria [3]. RLS sufferers were considered to be subjects who answered positively to all four questions, referring to the last month: (i) Do you have unpleasant sensations such as crawling or pain in your legs combined with an urge or need to move © 2014 The Author(s) European Journal of Neurology © 2014 EFNS
Association between RLS and migraine
your legs? (ii) Do these feelings/symptoms occur mainly or only at rest? (iii) Does movement improve these unpleasant sensations? (iv) Are these symptoms worse in the evening or at night than in the morning? RLS sufferers fulfilled the IRLSSG rating scale [22]. Covariate assessment
Depression was assessed by means of the Center for Epidemiologic Studies Depression (CES-D) scale [23] with a pathological score ≥23 suggesting the presence of major depression [24]. Sleep quality was assessed by the Pittsburgh Sleep Quality Index (PSQI) [25] considering as ‘poor sleep quality’ total scores of >5. Anxiety was assessed by the State-Trait Anxiety Inventory (STAI) scale [26]. CES-D, PSQI and STAI scales were self-administered. Possible secondary causes of RLS were also considered [3] evaluating the presence of low blood ferritin levels (ferritin
