Sage Journals: Discover world-class research

Abstract

Background

Restless legs syndrome (RLS) is increasingly being reported as a comorbidity of migraine.

Methods

We conducted a systematic review and meta-analysis of studies investigating RLS in headache/migraine and vice versa. We calculated the prevalence and 95% confidence intervals (CI) of RLS in headache/migraine, of headache/migraine in RLS and controls, and odds ratios (ORs) of the association between the conditions. We then determined pooled effect estimates for the associations.

Results

We identified 24 studies. RLS prevalence in migraine ranged from 8.7% to 39.0% with no apparent differences based on gender and aura status. Prevalence among controls was compatible with the literature. Migraine prevalence in RLS ranged from 15.1% to 62.6%. We did not pool prevalence data because of high unexplained heterogeneity. High heterogeneity with respect to the association between any migraine and RLS could be explained by study design. Pooled analyses showed substantially higher effect estimates in case-control studies (pooled OR = 4.19, 95% CI 3.07–5.71; I² = 0.0%) than in cohort studies (pooled OR = 1.22, 95% CI 1.14–1.30; I² = 0.0%).

Conclusion

Our results support the concept of RLS as an important comorbidity of migraine. However, the degree of association appears to be strongly determined by study design. Potential effects by gender and aura status and the role of RLS in other headache disorders remain unclear.

Keywords

Migraine restless legs syndrome systematic review meta-analysis prevalence association

Introduction

Migraine is a common, chronic neurological disorder affecting 10%–20% of the general population, predominantly women (1). It presents with recurrent headache attacks, associated vegetative symptoms, and hypersensitivity of various functional systems of the nervous system (1).

The comorbidity burden of migraine is high, and among the disorders most firmly associated with migraine are for example cardiovascular disease (CVD), in particular stroke (2), depression (3), and other pain disorders (4,5).

In addition, there is accumulating evidence that restless legs syndrome (RLS) is another condition frequently reported by migraine patients (6). RLS is a common sleep-related movement disorder that—similar to migraine—predominates in women (7). RLS is characterized by an urge to move the legs, often accompanied by unpleasant leg sensations. Symptoms typically occur at night and during rest and resolve during movement (8). A common origin for migraine and RLS has been proposed (9) and a pathophysiological link involving a disturbance of iron and dopamine metabolism may exist (6,10,11).

Over the past years a number of studies have been published investigating the association between headaches, in particular migraine, and RLS in different populations (12 –35). We therefore sought to summarize the current published evidence on RLS prevalence among migraine patients and the association between headaches, in particular migraine, and RLS by systematically reviewing the literature and performing a meta-analysis.

Subjects and methods

For this systematic review we followed the guidelines of the Meta-analysis of Observational Studies in Epidemiology (MOOSE) group for the design, performance, and reporting of meta-analyses of observational studies (36). We did not seek approval from an ethics committee since we used only data publicly available from the medical literature.

Data sources and searches

Two of the investigators (A.W., M.S.) independently searched PubMed, Embase, and Web of Science from their inceptions through January 2014. For their search the investigators combined text words and Medical Subject Headings (MESH) terms, if appropriate, for headache and migraine (“headache” or “headache disorders” or “migraine” or “migraine disorders”) with terms for restless legs syndrome (“restless legs syndrome” or “RLS”). They used the “explode” feature where applicable and restricted their search to humans. No language restrictions were applied. Additionally, the investigators manually searched the reference lists of all primary and review articles.

Study selection

A priori, we defined the following inclusion criteria for studies: First, we considered studies with a case-only (no control group), cross-sectional, case-control, or cohort design. Second, studies were required to investigate patients with migraine or headache or with RLS. Third, we considered only studies published after the publication of International Classification of Headache Disorders (ICHD)-I criteria for headache disorders (37) and the essential diagnostic criteria for RLS (38). Fourth, studies had to either provide prevalence estimates of RLS among migraine or headache patients or prevalence estimates of migraine or headache among RLS patients or relative effect estimates with 95% confidence intervals (CI) for the association of these conditions in the study populations.

For study selection two investigators (M.S., A.W.) jointly screened the titles and abstracts of all publications identified and excluded those not meeting any of the pre-specified criteria. The same investigators then evaluated the full articles of the remaining publications and also excluded any studies not meeting our inclusion criteria. Disagreements were resolved by consensus.

Data extraction

Two investigators (M.S., A.W.) independently extracted data from the studies identified and entered them into a database. Extracted data included the authors' names and title of the study, country of study origin, year of publication, study design, source population, study size, duration of follow-up if applicable, age and gender distribution of participants, criteria for diagnosing migraine/headache and RLS, temporal relationship between migraine/headache and RLS diagnosis, and use of migraine-specific drugs. For studies investigating RLS in migraine/headache patients we further extracted the number of individuals with and without migraine/headache, number of individuals with and without RLS among those with and without migraine/headache, and prevalence of RLS. For studies investigating migraine/headache among RLS patients we extracted the number of individuals with and without RLS, number of individuals with and without migraine/headache among those with and without RLS, and prevalence of migraine/headache. Further, we extracted the effect estimates with 95% CIs for the association between the two conditions. If effect estimates from various models were presented, we selected those derived from the model with the maximum amount of covariate acknowledgment. We did not contact study authors to obtain additional information.

Statistical analysis

To further investigate RLS prevalence, we calculated the percentage of RLS sufferers among migraine/headache patients and people without migraine/headache along with the 95% CIs if published data permitted to do so. For studies without a control group, we calculated only the percentage of RLS sufferers among migraine/headache patients. We then performed the DerSimonian and Laird Q test for heterogeneity among the study groups (migraine/headache patients, controls) and calculated the I² statistic (39). Because of high heterogeneity of RLS prevalence among the studies both in migraineurs and people without migraine (see Results section), we did not pool the prevalence data. We proceeded likewise to investigate the prevalence of headache/migraine in patients with RLS.

To further investigate the association between migraine/headache and RLS, we calculated the odds ratios (ORs) and 95% CIs for the association between migraine/headache and RLS in studies not providing relative effect estimates. We performed the DerSimonian and Laird Q test for heterogeneity and calculated the I² statistic (39). For meta-analysis we weighted the log of the ORs for each study by the inverse of their variance to obtain pooled estimates. We ran random-effects models, which include assumptions about potential differences between studies. Meta-regression was used to statistically evaluate the extent of heterogeneity due to country of study origin (Western populations vs. Asian populations vs. populations of heterogeneous ethnicities), study design (case-controls vs. cohort studies), gender, and migraine aura status. To visually examine the impact of individual studies on the overall homogeneity of the test statistic, we inspected Galbraith plots (40). Finally, to test for small study effects, e.g. due to publication bias, we used statistical methods described by Begg and Mazumdar (41) and Egger (42). Small study effects denote the tendency for smaller studies in meta-analysis to show larger effects.

A two-tailed p value<0.05 was considered statistically significant. All analyses were carried out using SAS v.9.2 (SAS Institute Inc, Cary, NC, USA) and STATA v.10.1 (Stata, College Station, TX, USA).

Results

Figure 1 summarizes the study selection process. The electronic search identified 752 references. After reviewing the titles and abstracts and excluding records not meeting our inclusion criteria, we were left with 27 records for further evaluation. We identified one additional record from electronic alerts for articles published online ahead of print (35). We evaluated these 28 records as full text references and excluded four that turned out to be conference abstracts not providing prevalence measures or measures of association between migraine/headache and RLS (43 –46). This left us with 24 studies for our systematic review (12 –35).

Figure 1.

Summary of the study selection process.

Study characteristics

The characteristics of the 24 studies are summarized in Table 1.

Table 1(a).

Characteristics of studies investigating RLS in headache/migraine.

Study, country	Publication type	Source population	Study design	Total study size	Mean age (years)	Gender	Headache/migraine diagnostic criteria	Headache categories (n, % women)	Average duration of headache/ migraine, years (±SD)	RLS diagnostic criteria	Time of RLS diagnosis relative to headache/migraine diagnosis	Headache/ migraine drug use
Young 2003, US (33)	Paper	Headache patients from an outpatient infusion center	Descriptive case only	50	37.9 ± 10.81	Women + men	NA	Headache (n = 50, 86%), any migraine (n = 41, NA), CM (n = 34, NA), EM (n = 7, NA), non-migraine headache (n = 9, NA), NDPH (n = 5, NA), PTH (n = 2, NA), CH (n = 2, NA)	NA	IRLSSG	NA	Yes
Rhode 2007, Germany (24)	Paper	Migraine: patients from headache outpatient clinic and “Dortmunder Gesundheitsstudie”; controls: waiting relatives of patients in trauma outpatient clinic and from “Dortmunder Gesundheitsstudie”	Case-control (matched by age and sex)	822	Controls: 42.1 ± 12.8; migraine: 41.7 ± 12.5	Women + men	ICHD-II	Controls (n = 411, 62.5%), any migraine (n = 411, 62.5%)	18.7 (±12.1)	Interview using structured questionnaire (IRLSSG criteria) confirmed by clinical diagnosis by a neurologist (lifetime prevalence)	NA (lifetime prevalence of RLS asked)	Yes
d'Onofrio 2008, Italy (17)	Paper	Headache: patients from three Italian headache centers; controls: hospital employees and visitors	Case-control (matched by age and sex)	320	Controls: 37.2 ± 8.5; headache: 37.2 ± 11.1	Women + men	ICHD-II (face-to-face interview)	Controls (n = 120, 75%), headache (n = 200, 74.5%), MA (n = 10, 50%), MO (n = 114, 78.9%), ETTH (n = 22, 72.7%), ECH (n = 12, 8.3%), PSH (n = 2, 100%), combinations (n = 40, 87.5%)	14.3 (±10.1)	Face-to-face interview (IRLSSG)	NA	No prophylactic medication or antidepressants for ≥3 months
Cologno 2008, Italy (13)	Paper	Migraine patients from Italian headache centers	Descriptive case-only	164	37.1 ± 10.8	Women + men	ICHD-II (face-to-face interview)	Any migraine (n = 164, 79.3%)	NA	Face-to-face interview (IRLSSG)	NA	No prophylactic medication
Çetīnkaya 2009, Turkey (34)	Paper	Patients attending a headache clinic	Descriptive case-only	76	49.64	NA	ICHD-II	TTH (n = 20, NA)	NA	IRLSSG	NA	NA
Chen 2010, Taiwan (12)	Paper	Patients attending the headache clinic of Taipei Veterans General Hospital	Descriptive case-only	1041	Any migraine: 42.1 ± 13.3; TTH: 50.6 ± 16.2; CH: 38.4 ± 12.6	Women + men	ICHD-II	Any migraine (n = 772, 83.4%), MA (n = 48, NA), MO (n = 267, NA), CM (n = 414, NA), PM (n = 43, NA), TTH (n = 218, 67.0%), CH (n = 51, 21.6%)	NA	Questionnaire (IRLSSG) followed by phone interview	NA	Yes, 15.3% of all patients
d'Onofrio 2011(a), Italy (18)	Paper	Migraine: patients from five Italian headache centers	Descriptive case-only	63	38.2 ± 14.1	Women + men	ICHD-II (face-to-face interview)	MA (n = 63, 52.4%)	NA	Face-to-face interview (IRLSSG)	NA	NA
d'Onofrio 2011(b), Italy (14)	Paper	CH: patients from two tertiary headache centers, controls: hospital employees and visitors free of headache	Case-control (matched by age and sex)	100	Controls: 39.3 ± 10.4; CH: 39.7 ± 10.9	Women + men	ICHD-II (face-to-face interview)	Controls (n = 50, 12%), CH (n = 50, 12.0%)	7.4 (±4.4)	IRLSSG	NA	yes, 26% of patients took preventative medications
Suzuki 2011, Japan (28)	Paper	Migraine: patients attending headache outpatient clinic; controls: recruited from hospital employees, their friends and family.	Case-control (matched by age and sex)	425	Controls: 37.3 ± 12.1; migraine: 38.2 ± 13.0	Women + men	ICHD-II (face-to-face interview)	Controls (n = 163, 84.7%), any migraine (n = 262, 82.1%)	NA	face-to-face interview (IRLSSG)	NA	yes, 50.8% of migraine patients
Kanki 2011, Japan (21)	Abstract	Patients attending a headache outpatient clinic	Descriptive case-only	255	NA	NA	NA	Headache (n = 255; NA), any migraine (n = 188, NA), MA (n = 24, NA), MO (n = 164, NA), nonmigraine headache (n = 67, NA)	NA	NA	NA	NA
Lucchesi 2012, Italy (23)	Paper	Migraine: patients attending the headache center at University of Pisa; controls: NA	Case-control (matched by age and sex)	477	Controls: 41.7 ± 14.2; migraine: 41.2 ± 13.5	Women + men	ICHD-II	Controls (n = 200, 68.5%), any migraine (n = 277, 68.6%), MA (n = 34, NA), MO (n = 243, NA), EM (n = 175, NA), CM (n = 102, NA)	NA	IRLSSG	NA	No preventives and antidepressants during the preceding three months
Seidel 2012, Austria (27)	Paper	Cases: patients attending the Headache Unit at the Department of Child and Adolescent Psychiatry (University of Vienna); control group 1: patients from Department of Pediatrics and Adolescent Medicine after recovery; (control group 2: children and adolescents from primary and grammar schools)	Case-control	292	Control group 1: 13 ± 5, control group 2: 13 ± 4, migraine: 13 ± 3	Girls + boys	ICHD-II (semi-structured interview)	Controls 1 (n = 73, 54.8%), controls 2 (n = 108, 54.6%), any migraine (n = 111, 50.5%).	3.2 (±2.8)	IRLSSG (asking about previous six months)	NA (RLS questions cover previous six months)	No prophylactic medication, no metoclopramide
Karthik 2012, India (22)	Paper	Cases: neuropsychiatric patients attending a tertiary care university center; controls: participants of a study to detect prevalence of sleep abnormalities in the general population	Case-control (matched by age and sex)	180	Controls: 31.76 ± 8.2; cases: 31.76 ± 8.2	Women + men	ICHD-II	Controls (n = 90; 78.9%), MO (n = 90; 78.9%)	5.9 (±6.1)	NA	NA	No use of migraine prophylaxis
De Martino 2012, Italy (15)	Abstract	Patients attending outpatient clinic of a headache center	Descriptive case-only	136	53.52 ± 12.0	Women + men	ICHD-II	Any migraine (n = 136, 75%)	NA	IRLSSG	NA	NA
Rozen 2012, US (25)	Paper	CH patients participating in the web-based US Cluster Headache Survey	Descriptive case-only	1134	NA	Women + men	Self-report	CH (n = 1134, 28%)	NA	Self-report	NA	NA
Schürks 2012, US (26)	Paper	Women's Health Study	Cohort	31,370	No migraine: 63.9 ± 7.1; any migraine: 62.5 ± 6.3 (at time of RLS diagnosis)	Women	Modified ICHD-I (questionnaire)	Controls (n = 24,513, 100%), any migraine (n = 6857, 100%), MA (n = 1579, 100%), MO (n = 2418, 100%), prior migraine (n = 1725, 100%), new reports of migraine (n = 1135, 100%)	NA	Self-administered questionnaire (IRLSSG)	Migraine was reported at time of or before RLS diagnosis	NA
Winter 2013, US (32)	Paper	Physicians' Health Study	Cohort	22,926	67.8 ± 8.9 (at time of RLS diagnosis)	Men	Self-report on questionnaire	Controls (n = 20,110, 0%), any migraine (n = 2816, 0%)	NA	Self-administered questionnaire (IRLSSG)	Migraine was reported at time of or before RLS diagnosis	NA
Ferreira 2013, Brazil (19)	Paper	Employees of University Hospital, São Paulo	Case-control (matched by age, sex, and employment type)	144	Controls: 40; migraine: 42.5	Women + men	ICHD-II (face-to-face interview)	Controls (n = 72, 79.2%), any migraine (n = 72, 86.1%)	NA	Questionnaire-based interview (IRLSSG)	NA	NA

Table 1(b).

Characteristics of studies investigating headache/migraine in RLS.

Study, country	Publication type	Source population	Study design	Total study size	Mean age (years)	Gender	Headache/ migraine diagnostic criteria	RLS categories (n, % women)	Average duration of RLS, years (±SD)	RLS diagnostic criteria	Time of RLS diagnosis relative to headache/migraine diagnosis	RLS-specific treatment
Ulfberg 2001(a), Sweden (29)	Paper	Random sample of 4000 men aged 18–64 from the general Swedish population	Cross-sectional	2608	Men with RLS: 47	Men	NA	No RLS (n = 2377, 0%), RLS (n = 231, 0%)	NA	IRLSSG	NA	NA
Ulfberg 2001(b), Sweden (30)	Paper	Random sample of 200 women aged 18–64 from the general Swedish population	Cross-sectional	140	Women with RLS: 45	Women	NA	No RLS (n = 124, 100%), RLS (n = 16, 100%)	NA	IRLSSG	NA	NA
Wesström 2008, Sweden (31)	Paper	Random sample of 5000 women aged 18–64 from the general Swedish population	Cross-sectional	3516	NA	Women	NA	No RLS (n = 2950, 100%), RLS (n = 551, 100%)	NA	IRLSSG	NA	NA
Gupta 2012, India (20)	Paper	Patients attending a sleep clinic in a tertiary care teaching institution	Descriptive case-only	99	39.91	Women + men	ICHD-II	Primary RLS (n = 99, 67.7%)	4.39 ± 5.68	AASM	NA	NA
Dickoff 2013, US (16)	Abstract	NA	Descriptive case-only	700	NA	NA	NA	RLS (n = 700, NA)	NA	IRLSSG	NA	NA
Gozubatik-Celik 2014, Turkey (35)	Paper	Patients attending the Sleep Disorder Unit at University Neurology Department (University of Istanbul)	Descriptive case-only	265	50.4 ± 12.8	Women + men	ICHD-II	Primary RLS (n = 265, 48.3%)	8.40 ± 8.6	IRLSSG	NA	Yes, 58.1% of patients

RLS: restless legs syndrome; US: United States; IRLSSG: International Restless Legs Syndrome Study Group; CM: chronic migraine; EM: episodic migraine; MA: migraine with aura; MO: migraine without aura; PM: probable migraine; NDPH: new daily persistent headache; PTH: post-traumatic headache; CH: cluster headache; ECH: episodic cluster headache; TTH, tension-type headache; ETTH: episodic tension-type headache; PSH: primary stabbing headache; ICHD-II: International Classification of Headache Disorders, second edition; AASM: American Academy of Sleep Medicine; NA: not available.

Eighteen studies reported on RLS among migraine/headache patients (12 –15,17 –19,21 –28,32 –34), among those 16 are full papers (12 –14,17 –19,22 –28,32 –34) and two are abstracts (15,21). One study was in a pediatric and adolescent population (27), while the other studies were in adults. Two studies had a cohort design (26,32) and eight studies each had a case-control (14,17,19,22 –24,27,28) or descriptive case-only design (12,13,15,18,21,25,33,34). In 15 studies migraine patients were investigated (12,13,15,17 –19,21 –24,26 –28,32,33), four studies reported on cluster headache (CH) patients (12,14,25,33), and five studies looked at mixed headache forms and other headache disorders (12,17,21,33,34).

Six studies reported on migraine/headache among patients with RLS (16,20,29 –31,35); among those five are full papers (20,29 –31,35) and one is an abstract (16). Three studies each had a cross-sectional (29 –31) or a descriptive case-only design (16,20,35). Further, three studies each reported on migraine (16,20,35) or non-ICHD defined headaches (morning/daytime headaches) (29 –31) among RLS patients.

Prevalence of RLS in migraine/headache

Eighteen studies provided information allowing calculation of RLS prevalence among migraine/headache patients (Table 2) (12 –15,17 –19,21 –28,32 –34). However, only 10 studies provided data from people without migraine/headache (14,17,19,22 –24,26 –28,32).

Table 2.

Prevalence of RLS among headache/migraine patients and controls and association between both conditions from individual studies.

Study, year	Gender	Study population (n, % women)	RLS present, n	RLS absent, n	Prevalence^a (95% CI)	Mean severity (IRLSSG)	OR^b (95% CI)
Young, 2003 (33)	Women + men	Headache (n = 50, 86%)	17	33	34.00 (20.40–47.60)	NA	NA
		Any migraine (n = 41, NA)	16	25	39.02 (23.44–54.61)	NA	NA
		CM (n = 34)	13	21	38.24 (21.02–55.45)	NA	NA
		EM (n = 7)	3	4	42.86 (–6.58 to 92.29)	NA	NA
		Nonmigraine headache (n = 9, NA)	1	8	11.11 (–14.51 to 36.73)	NA	NA
		NDPH (n = 5)	1	4	20.00 (–35.53 to 75.53)	NA	NA
		PTH (n = 2)	0	2	0	NA	NA
		CH (n = 2)	0	2	0	NA	NA
Rhode, 2007 (24)	Women + men	Controls (n = 411, 62.5%)	23	388	5.60 (3.37–7.83)	10.0 ± 2.0	Referent
		Any migraine (n = 411, 62.5%)	71	340	17.28 (13.61–20.95)	15.6 ± 6.6	3.5 (2.2–5.8)
d'Onofrio, 2008 (17)	Women + men	Controls (n = 120, 75%)	10	110	8.33 (3.32–13.35)	NA	Referent
		Headache (n = 200, 74.5%)	44	156	22.00 (16.21–27.79)	NA	3.10 (1.50–6.43)
Cologno, 2008 (13)	Women + men	Any migraine (n = 164, 79.3%)	42	122	25.61 (18.86–32.36)	NA	NA
		Women (n = 130, 100%)	35	95	26.92 (19.20–34.65)	NA	NA
		Men (n = 34, 0%)	7	27	20.59 (6.27–34.91)	NA	NA
Çetīnkaya, 2009 (34)	NA	TTH (n = 20, NA)	1	19	5.00 (–5.47 to 15.47)	NA	NA
Chen, 2010 (12)	Women + men	Any migraine (n = 772, 83.4%)	88	684	11.40 (9.15–13.65)	16.7 ± 7.9	NA
		Women (n = 644, 100%)	76	568	11.80 (9.30–14.30)	NA	NA
		Men (n = 128, 0%)	12	116	9.38 (4.26–14.49)	NA	NA
		MA (n = 48, NA)	4	44	8.33 (0.22–16.44)	NA	NA
		MO (n = 267, NA)	29	238	10.86 (7.11–14.62)	NA	NA
		CM (n = 414, NA)	49	365	11.84 (8.71–14.96)	NA	NA
		PM (n = 43, NA)	6	37	13.95 (3.16–24.74)	NA	NA
		TTH (n = 218, 67.0%)	10	208	4.59 (1.79–7.39)	15.1 ± 3.2	NA
		CH (n = 51, 21.6%)	1	50	1.96 (–0.20 to 5.90)	13	NA
d'Onofrio, 2011(a) (18)	Women + men	MA (n = 63, 52.4%)	6	57	9.52 (2.07–16.98)	NA	NA
d'Onofrio, 2011(b) (14)	Women + men	Controls (n = 50, 12.0%)	6	44	12.00 (2.67–21.33)	NA	Referent
		CH (n = 50, 12.0%)	0	50	0	NA	(—)
Suzuki, 2011 (28)	Women + men	Controls (n = 163, 84.7%)	3	160	1.84 (–0.25 to 3.93)	6	Referent
		Any migraine (n = 262, 82.1%)	36	226	13.74 (9.54–17.94)	14	8.50 (2.57–28.07)
Kanki, 2011 (21)	NA	Headache (n = 255; NA)	52	203	20.39 (15.41–25.37)	NA	NA
		Any migraine (n = 188, NA)	40	148	21.28 (15.37–27.18)	NA	NA
		MA (n = 24, NA)	8	16	33.33 (13.00–53.67)	NA	NA
		MO (n = 164, NA)	32	132	19.51 (13.38–25.64)	NA	NA
		Nonmigraine headache (n = 67, NA)	12	55	17.91 (8.49–27.33)	NA	NA
Lucchesi, 2012 (23)	Women + men	Controls (n = 200, 68.5%)	15	185	7.50 (3.82–11.18)	NA	Referent
		Any migraine (n = 277, 68.6%)	63	214	22.74 (17.78–27.71)	NA	3.63 (2.00–6.59)
		MA (n = 34, NA)	8	26	23.53 (8.51–38.55)	NA	3.80 (1.47–9.83)
		MO (n = 243, NA)	55	188	22.63 (17.34–27.93)	NA	3.61 (1.97–6.61)
		EM (n = 175, NA)	28	147	16.00 (10.52–21.49)	NA	2.35 (1.21–4.56)
		CM (n = 102, NA)	35	67	34.31 (24.94–43.69)	NA	6.44 (3.31–12.54)
	Women	Controls (n = 137, 100%)	11	126	8.03 (3.42–12.64)	NA	Referent
		Any migraine (n = 190, 100%)	47	143	24.74 (18.55–30.93)	NA	3.77 (1.87–7.57)
	Men	Controls (n = 63, 0%)	4	59	6.35 (0.16–12.54)	NA	Referent
		Any migraine (n = 87, 0%)	16	71	18.39 (10.09–26.70)	NA	3.32 (1.05–10.49)
Seidel, 2012 (27)	Girls + boys	Controls 1 (n = 73, 54.8%)	4	69	5.48 (0.13–10.83)	NA	Referent
		Controls 2 (n = 108, 54.6%)	9	99	8.33 (3.04–13.63)	NA	(—)
		Any migraine (n = 111, 50.5%)	24	87	21.62 (13.84–29.40)	NA	5.3 (1.5–18.6)
Karthik, 2012 (22)	Women + men	Controls (n = 90; 78.9%)	5	85	5.56 (0.73–10.38)	NA	Referent
		MO (n = 90; 78.9%)	29	61	32.22 (22.38–42.07)	NA	8.08 (2.96–22.05)
De Martino, 2012 (15)	Women + men	Any migraine (n = 136, 75%)	44	92	32.35 (24.39–40.32)	NA	NA
Rozen, 2012 (25)	Women + men	CH (n = 1134, 28%)	125	1009	11.02 (9.20–12.85)	NA	NA
Schürks, 2012 (26)	Women	Controls (n = 24,513, 100%)	2749	21,764	11.21 (10.82–11.61)	NA	Referent
		Any migraine (n = 6857, 100%)	996	5861	14.53 (13.69–15.36)	NA	1.22 (1.13–1.32)
		Controls (n = 24,434, 100%)	2738	21,696	11.21 (10.81–11.60)	NA	Referent
		MA (n = 1579, 100%)	239	1340	15.14 (13.37–16.91)	NA	1.27 (1.10–1.48)
		MO (n = 2418, 100%)	351	2067	14.52 (13.11–15.92)	NA	1.24 (1.09–1.40)
		Prior migraine (n = 1725, 100%)	232	1493	13.45 (11.84–15.06)	NA	1.11 (0.96–1.28)
		New reports of migraine (n = 1135, 100%)	174	961	15.33 (13.23–17.43)	NA	1.30 (1.10–1.54)
Winter, 2013 (32)	Men	Controls (n = 20,110, 0%)	1427	18,683	7.10 (6.74–7.45)	NA	Referent
		Any migraine (n = 2816, 0%)	245	2571	8.70 (7.66–9.74)	NA	1.20 (1.04–1.38)
Ferreira, 2013 (19)	Women + men	Controls (n = 72, 79.2%)	6	66	8.33 (1.79–14.87)	15	Referent
		Any migraine (n = 72, 86.1%)	18	54	25.00 (14.75–35.25)	18	3.67 (1.36–9.88)

Prevalence in percentage and 95% confidence intervals (CIs) were calculated based on the numbers of individuals with and without RLS among headache/migraine patients and controls as provided in the publications.

Odds ratios (ORs) and 95% CIs were extracted from the papers or calculated if the publications provided the numbers of individuals with and without RLS among headache/migraine patients and controls. RLS: restless legs syndrome; IRLSSG: International Restless Legs Syndrome Study Group; CM: chronic migraine; EM: episodic migraine; MA: migraine with aura; MO: migraine without aura; PM: probable migraine; NDPH: new daily persistent headache; PTH: post-traumatic headache; CH: cluster headache; ECH: episodic cluster headache; TTH: tension-type headache; ETTH: episodic tension-type headache; PSH: primary stabbing headache; NA: not available.

For any migraine the prevalence ranged from 8.7% (95% CI 7.7–9.7) (32) to 39.0% (95% CI 23.4–54.6) (33). The ranges among men (from 8.7%, 95% CI 7.7–9.7 (32) to 20.6%, 95% CI 6.3–34.9 (13)) and among women (from 11.8%, 95% CI 9.3–14.3 (12) to 26.9%, 95% CI 19.2–34.7 (13)) were largely overlapping. Likewise, the ranges for migraine with aura (from 8.3%, 95% CI 0.2–16.4 (12) to 33.3%, 95% CI 13.0–53.7 (21)) and migraine without aura (from 10.9%, 95% CI 7.1–14.6 (12) to 32.2%, 95% CI 22.4–42.1 (22)) were very similar.

The prevalence of RLS among patients with CH ranged from 0% (14,33) to 11.0% (95% CI 9.2–12.9) (25).

The two studies investigating patients with tension-type headache (TTH) reported a prevalence of 4.6% (95% CI 1.8–7.4) (12) and 5.0% (95% CI –5.5 to 15.5) (34).

Among controls of non-Asian populations 5.6% (95% CI 3.4–7.8) to 12.0% (95% CI 2.7–21.3) had RLS (14,17,19,23,24,26,27,32), compared with 1.8% (95% CI –0.3 to 3.9) to 5.6% (95% CI 0.7–10.4) among controls in Asian populations (22,28).

Heterogeneity was high among studies investigating RLS prevalence in patients with any migraine (I² = 92.8%; test for heterogeneity chi-square = 180.53, degree of freedom (df) = 13, p < 0.001), in CH (I² = 94.0%; test for heterogeneity chi-square = 16.71, df = 1, p < 0.001; two of the four studies were not included since none of the patients had RLS), and in controls (I² = 96.8%; test for heterogeneity chi-square = 284.3, df = 9, p < 0.001). Among studies looking at any migraine meta-regression did not indicate that country of study origin (Western (reference) vs. Asian (p = 0.82) vs. heterogeneous populations (p = 0.65)), study design (case only (reference) vs. case control (p = 0.98) vs. cohort (p = 0.15)), migraine aura status (p = 0.46) or gender (p = 0.26) account for significant heterogeneity across those studies. Likewise among studies reporting on RLS in controls meta-regression did not indicate that country of study origin (Western (reference) vs. Asian (p = 0.06) vs. heterogeneous populations (p = 0.94)) or design (p = 0.14) were significant sources of heterogeneity. Hence, we did not attempt to pool the data for any migraine, CH, and controls for meta-analysis.

Among the two studies investigating RLS prevalence in patients with TTH, there was no indication for heterogeneity (I² = 0.0%; test for heterogeneity chi-square = 0.01, df = 1, p = 0.94). Pooling the data indicated a prevalence of 4.6% (95% CI 1.9–7.3).

Prevalence of migraine/headache in RLS

Among the six studies reporting on migraine/headache in patients with RLS (Table 3) (16,20,29 –31,35) three reported on migraine (16,20,35). Among those the prevalence ranges from 15.1% (95% CI 10.7–19.4) (35) to 62.6% (95% CI 59.0–66.2) (16).

Table 3.

Prevalence of migraine among RLS patients and controls and association between both conditions from individual studies.

Study, year	Study population (n, % women)	Migraine present, n	Migraine absent, n	Prevalence (95% CI)^a	Mean severity (IRLSSG)	OR (95% CI)^b
Ulfberg 2001(a) (29)	No RLS (n = 2377, 0%)	NA (morning/ daytime headache)	NA (morning/daytime headache)	NA	NA	Referent
	RLS (n = 231, 0%)	NA (morning/ daytime headache)	NA (morning/daytime headache)	NA	NA	Morning headache: 4.7 (2.9–7.6); daytime headache: 2.8 (1.8–4.2)
Ulfberg 2001(b) (30)	No RLS (n = 124, 100%)	NA (daytime headache)	NA (daytime headache)	NA	NA	Referent
	RLS (n = 16, 100%)	NA (daytime headache)	NA (daytime headache)	NA	NA	4.8 (1.4–16.0)
Wesström 2008 (31)	No RLS (n = 2950, 100%)	182 (Morning headaches)	2768	6.17 (5.30–7.04)	NA	Referent
	RLS (n = 551, 100%)	69 (Morning headaches)	482	12.52 (9.75–15.30)	NA	2.07 (1.53–2.80)
Gupta 2012 (20)	Primary RLS (n = 99, 67.7%)	44	55	44.44 (34.48–54.41)	NA	NA
Dickoff 2013 (16)	RLS (n = 700, NA)	438	262	62.57 (58.98–66.17)	NA	NA
Gozubatik-Celik 2014 (35)	Primary RLS (n = 265, 48.3%)	40	225	15.09 (10.76–19.43)	19.7 ± 8.0	NA
	Women (n = 128, 100%)	13	115	10.16 (4.85–15.46)	NA	NA
	men (n = 137, 0%)	27	110	19.71 (12.96–26.45)	NA	NA

Heterogeneity among the studies investigating migraine prevalence in RLS patients was high (I² = 99.3%; test for heterogeneity chi-square = 272.43, df = 2, p < 0.001), hence, we did not pool the data for meta-analysis.

Association between migraine/headache and RLS

A total of 10 studies reported on RLS among migraine/headache patients and controls (14,17,19,22 –24,26 –28,32). Among those effect estimates for the association between any migraine and RLS were available for eight studies (Table 2) (19,22 –24,26 –28,32). ORs (95% CIs) ranged from 1.20 (1.04–1.38) (32) to 8.50 (2.57–28.07) (28). One additional study investigated the association between headache and RLS, presenting an OR of 3.10 (95% CI 1.50–6.43) (17). One study among patients with CH did not allow calculating an effect estimate since none of the CH patients had RLS (14).

Heterogeneity was high among studies investigating the association between any migraine and RLS (I² = 88.8%; test for heterogeneity chi-square = 62.32, df = 7, p < 0.001). Meta-regression did not indicate that country of study origin (Western (reference) vs. Asian (p = 0.09) vs. heterogeneous populations (p = 0.55)), gender (p = 0.87) or migraine aura status (p = 0.66) account for significant heterogeneity among those studies. However, meta-regression suggested that study design (p < 0.001) was a significant source of heterogeneity among the studies. This agreed with visual inspection of the Galbraith plots, where studies with a case-control design (19,22 –24,27,28) were located outside the margin set by two standard deviations of the z score. Hence, we did not attempt to pool the data from all studies for meta-analysis, but performed an analysis stratified by study design.

Results from the pooled analysis of case-control studies suggest that any migraine is associated with four-fold increased odds for RLS (pooled OR = 4.19, 95% CI 3.07–5.71), while pooled analysis of cohort studies suggest that the odds for RLS among migraineurs are increased by about 20% (pooled OR = 1.22, 95% CI 1.14–1.30) (Table 4, Figures 2 and 3). There was no remaining heterogeneity within the strata (case-control studies: I² = 0.0%, test for heterogeneity chi-square = 3.94, df = 5, p = 0.56; cohort studies: I² = 0.0%, test for heterogeneity chi-square = 0.04, df = 1, p = 0.84). There was no evidence for small study effects among case-control or cohort studies in formal investigations with Begg's and Egger's tests (Table 4).

Figure 2.

Results from the case-control studies.

Figure 3.

Results from the cohort studies.

Table 4.

Association between any migraine and RLS, heterogeneity, and small study effects.

			Heterogeneity				Small study effects p value
	No. of studies	Pooled OR (95% CI)^a	Q	df	p value	I² in %	Begg's test	Egger's test
Case control studies (19,22 –24,27,28)	6	4.19 (3.07–5.71)	3.94	5	0.56	0.0	0.09	0.07
Cohort studies (26,32)	2	1.22 (1.14–1.30)	0.04	1	0.84	0.0	0.32	–

From random-effects model. RLS: restless legs syndrome; OR: odds ratio; df: degree of freedom; CI: confidence interval.

The three studies reporting on migraine in RLS patients did not have a control group, obviating calculating an effect estimate.

Discussion

Analyses from this systematic review suggest that RLS prevalence for any migraine ranges from 8.7% to 39.0%, with similar ranges for women and men as well as for migraine with aura and migraine without aura. Among controls of non-Asian populations 5.6% to 12.0% had RLS, compared with 1.8% to 5.6% among controls in Asian populations. Heterogeneity with respect to prevalence was high both among studies in patients with any migraine and controls. Prevalence of migraine in RLS ranges from 15.1% to 62.6%; heterogeneity was likewise high. Data regarding prevalence of RLS in CH and TTH as well as prevalence of migraine in RLS are limited. All published studies report increased odds for RLS among patients with any migraine. Heterogeneity among those studies was high and is attributable to study design. Among case-control studies pooled analysis indicates that any migraine is associated with four-fold increased odds for RLS, while pooled analysis of cohort studies suggest that the odds for RLS among migraineurs are increased by only about 20%.

There is plausibility for an association between migraine and RLS when considering pathophysiology and clinical findings. In both conditions disturbances of the iron and dopamine metabolism have been indicated. First, brain iron deposition has been reported in migraine patients and increased iron accumulation appears to be associated with repeated attacks (47). On the contrary, iron deficiency has been implicated in at least a subset of patients with RLS (48). Second, dopamine has long been hypothesized to be involved in migraine pathophysiology (10). In particular, premonitory symptoms like yawning, irritability, and mood changes as well as nausea and vomiting occurring both during the premonitory and headache phases may be caused by dopamine. Further, antiemetics with antidopaminergic properties are effective in aborting migraine attacks (49). These findings agree with experimental data from rats that dopamine can modulate neuronal firing in the trigeminocervical complex (50) and hence, modulate pain transmission in the brainstem. With regard to RLS pathophysiology, the most widely supported hypothesis relates to a dysfunction of the dopaminergic system of the A11 neurons descending from the hypothalamus to the spinal cord (11). A “dopaminergic link” between migraine and RLS is further supported by reports that migraineurs with RLS more often have premonitory symptoms than migraineurs without RLS (13).

Available data favor the scenario that RLS is a prevalent comorbidity in migraine and RLS is a condition that may substantially impair quality of life (51). However, for the following reasons, the burden of RLS imposed on migraine patients is still difficult to quantify. First, the reported RLS prevalence varies greatly among studies (Table 2), resulting in high heterogeneity. Our analyses do not indicate that country of study origin, study design, migraine aura status or gender significantly account for this heterogeneity. Second, many of the studies do not have a control group, and among studies with control groups heterogeneity in the control groups was likewise high (Table 2). This could not be explained to a significant degree by design and country of study origin. The observed prevalence range among non-Asian control populations of 5.6% to 12.0% was within the range of 4%–29% reported in a recent systematic review (7). The prevalence range of 1.8%–5.6% among Asian controls was lower than among the non-Asian controls, which agrees with the published literature (52); however, interpretation warrants caution because of the low number of studies. Hence there are remaining uncertainties with respect to RLS prevalence among migraineurs.

The findings among controls argue that the wide range in RLS prevalence among patients with any migraine cannot be ascribed to migraine per se. A potential source of variability is frequency and severity of RLS symptoms used to aid RLS diagnosis. However, only four studies reported some form of International Restless Legs Syndrome Study Group (IRLSSG) scores (Table 1). Hence, too few studies are available that could be sensibly grouped into categories to be used in a meta-regression approach. Additional factors such as genetics and environmental factors also likely account for the variability of RLS prevalence. This is supported by reports that the risk for RLS increases with multimorbidity, i.e. appears to be a function cumulative disease burden (53).

High heterogeneity among the studies with respect to the effect estimates was completely explained by study design. Specifically studies with a case-control design suggested four-fold increased odds of RLS for any migraine, while among cohort studies there were only 20% increased odds, with no remaining heterogeneity. Hence, when interpreting our results, the following aspects need to be considered.

First, despite the established criteria by the International Headache Society (54) migraine is still clinically very heterogeneous. Hence, migraineurs presenting to headache or neurology clinics, where cases for the case-control studies were recruited, are likely more seriously affected by their migraines than migraineurs identified in non-clinic cohort studies. If we further hypothesize that migraine severity is correlated with comorbidity including RLS, it is plausible that the odds for RLS are higher in case-control than cohort studies. This may be supported by data from one cohort study, suggesting that active migraine at the time of the study is associated with RLS, while such an association was no longer statistically significant for prior migraine, i.e. for women no longer reporting active migraine at the time of the study (26). However, data on migraine attack frequency, an important marker for migraine severity, were not available in this study.

Second, patients and controls in case-control studies were asked about their migraines and RLS without taking the temporal association between the two conditions into consideration, while the cohort studies were designed to investigate only RLS among participants who had reported migraine previously. Hence, the effect estimates reported in case-control studies are more likely to reflect the overall risk of having the two conditions together (not necessarily at the same time or in a certain sequence), while the cohort studies estimate the future risk of RLS among migraineurs. In both types of study designs RLS was ascertained at only one point in time obviating analyses of changing time patterns of RLS manifestation.

Third, the impact of potential confounders on the association between migraine and RLS was differentially captured among studies. While some of the case-control studies were matched by age and sex (17,19,22 –24,28), which are established risk factors both for migraine and RLS, a number of case-control studies do not report if covariate acknowledgment was taken into account. In contrast the cohort studies additionally accounted for many covariates that may be associated with migraine and/or RLS and as such affect the respective other condition, including for example cardiovascular risk factors, alcohol consumption, depression, and Parkinson's disease (26,32).

Finally, use of migraine-preventive drugs may affect RLS prevalence among migraineurs; however, available data did not permit us to investigate this.

Our analyses support the role of RLS as an important comorbidity in migraine and an association between both conditions. The degree of association may be determined by severity of migraine. In addition, defining the temporal sequence for the occurrence of migraine and RLS as well as taking changing time patterns for both conditions into consideration might impact their association. The available data, although scarce, do not support a potential modifying effect by gender and migraine aura status. Further, a potential association between RLS and other primary headache disorders like CH and TTH needs to be further investigated. Given the high prevalence and burden of both headache disorders and RLS in the general population, future studies addressing these unresolved questions are of great individual and public health importance.

Clinical implications

Migraine and restless legs syndrome (RLS) are common neurological conditions in clinical practice.

RLS is an important comorbidity of migraine.

The propensity of having RLS appears to be higher among patients in a clinic-based setting compared to migraineurs from the general population.

Patients presenting with either migraine or RLS should also be questioned about the respective other conditions, since this may have implications for treatment strategy.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

We present full disclosures for all authors. None of the disclosures for any of the authors represent a conflict of interest with regard to this specific manuscript.

Dr Schürks is a full-time employee of Bayer HealthCare Germany.

Dr Winter is currently supported by funds from Washington University School of Medicine, the Barnes-Jewish Hospital Foundation, and Siteman Cancer Center.

Dr Berger has received research support for the conduction of the DMKG Headache Study; unrestricted grants of equal share from the German Migraine and Headache Society and a consortium formed by Allmiral, Astra-Zeneca, Berlin-Chemie, Boehringer Ingelheim Pharma, Boots Healthcare, GlaxoSmithKline, Janssen Cilag, McNeil Pharmaceuticals, MSD Sharp & Dohme, Pfizer; for the “Course of Restless Legs Syndrome Study”; unrestricted grants from the German Restless Legs Society and a consortium formed by Boehringer Ingelheim Pharma, Mundipharma Research, Neurobiotec, UCB (Schwarz Pharma) and Roche Pharma; from the German Ministry of Research and Education for several research projects within the German Competence Net Stroke and an ongoing cohort study on depression and subclinical arteriosclerosis.

Dr Kurth has received within the last two years investigator-initiated research funding from the French National Research Agency and the US National Institutes of Health. Further, he has received honoraria from BMJ and Cephalalgia for editorial services.

References

Haut

Bigal

Lipton

. Chronic disorders with episodic manifestations: Focus on epilepsy and migraine. Lancet Neurol 2006; 5: 148–157.

Schürks

Rist

Bigal

. Migraine and cardiovascular disease: A systematic review and meta-analysis. BMJ 2009; 339: b3914.

Rist

Schürks

Buring

. Migraine, headache, and the risk of depression: Prospective cohort study. Cephalalgia 2013; 33: 1017–1025.

Scher

Stewart

Lipton

. The comorbidity of headache with other pain syndromes. Headache 2006; 46: 1416–1423.

Tietjen

Brandes

Peterlin

. Allodynia in migraine: Association with comorbid pain conditions. Headache 2009; 49: 1333–1344.

Cannon

Larner

. Migraine and restless legs syndrome: Is there an association? J Headache Pain 2011; 12: 405–409.

Innes

Selfe

Agarwal

. Prevalence of restless legs syndrome in North American and Western European populations: A systematic review. Sleep Med 2011; 12: 623–634.

Allen

Picchietti

Hening

. 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: 101–119.

Sabayan

Bagheri

Borhani Haghighi

. Possible joint origin of restless leg syndrome (RLS) and migraine. Med Hypotheses 2007; 69: 64–66.

10.

Akerman

Goadsby

. Dopamine and migraine: Biology and clinical implications. Cephalalgia 2007; 27: 1308–1314.

11.

Paulus

Dowling

Rijsman

. Pathophysiological concepts of restless legs syndrome. Mov Disord 2007; 22: 1451–1456.

12.

Chen

Fuh

Chen

. Association between restless legs syndrome and migraine. J Neurol Neurosurg Psychiatry 2010; 81: 524–528.

13.

Cologno

Cicarelli

Petretta

. High prevalence of dopaminergic premonitory symptoms in migraine patients with restless legs syndrome: A pathogenetic link? Neurol Sci 2008; 29(Suppl 1): S166–S168.

14.

d'Onofrio

Cologno

Cicarelli

. Cluster headache patients are not affected by restless legs syndrome: An observational study. Clin Neurol Neurosurg 2011; 13: 308–310.

15.

De Martino

Peila

Rainero

. Migraine and restless legs syndrome: Further investigating the link. Neurology 2012; 78(Suppl): 04249.

16.

Dickoff

Dontje

. Restless limbs syndrome (RLS), migraine and bruxism: A common clinical triad. Ann Neurol 2013; 74(Suppl 17): S47.

17.

d'Onofrio

Bussone

Cologno

. Restless legs syndrome and primary headaches: A clinical study. Neurol Sci 2008; 29(Suppl 1): S169–S172.

18.

d'Onofrio

Cologno

Petretta

. Restless legs syndrome is not associated with migraine with aura: A clinical study. Neurol Sci 2011; 32(Suppl 1): S153–S156.

19.

Ferreira

Eckeli

Dach

. Comorbidities, medications and depressive symptoms in patients with restless legs syndrome and migraine. Arq Neuropsiquiatr 2013; 71: 87–91.

20.

Gupta

Lahan

Goel

. Primary headaches in restless legs syndrome patients. Ann Indian Acad Neurol 2012; 15: S104–S108.

21.

Kanki

Takeshima

. Restless legs syndrome as a comorbidity of migraine. Cephalalgia 2011; 31(Suppl 1): 160.

22.

Karthik

Kulkarni

Taly

. Sleep disturbances in ‘migraine without aura’—a questionnaire based study. J Neurol Sci 2012; 321: 73–76.

23.

Lucchesi

Bonanni

Maestri

. Evidence of increased restless legs syndrome occurrence in chronic and highly disabling migraine. Funct Neurol 2012; 27: 91–94.

24.

Rhode

Hosing

Happe

. Comorbidity of migraine and restless legs syndrome—a case-control study. Cephalalgia 2007; 27: 1255–1260.

25.

Rozen

Fishman

. Cluster headache in the United States of America: Demographics clinical characteristics triggers, suicidality, and personal burden. Headache 2012; 52: 99–113.

26.

Schürks

Winter

Berger

. Migraine and restless legs syndrome in women. Cephalalgia 2012; 32: 382–389.

27.

Seidel

Bock

Schlegel

. Increased RLS prevalence in children and adolescents with migraine: A case-control study. Cephalalgia 2012; 32: 693–699.

28.

Suzuki

Miyamoto

. Evaluation of contributing factors to restless legs syndrome in migraine patients. J Neurol 2011; 258: 2026–2035.

29.

Ulfberg

Nystrom

Carter

. Prevalence of restless legs syndrome among men aged 18 to 64 years: An association with somatic disease and neuropsychiatric symptoms. Mov Disord 2001; 16: 1159–1163.

30.

Ulfberg

Nystrom

Carter

. Restless legs syndrome among working-aged women. Eur Neurol 2001; 46: 17–19.

31.

Wesström

Nilsson

Sundstrom-Poromaa

. Restless legs syndrome among women: Prevalence, co-morbidity and possible relationship to menopause. Climacteric 2008; 11: 422–428.

32.

Winter

Schürks

Berger

. Migraine and restless legs syndrome in men. Cephalalgia 2013; 33: 130–135.

33.

Young

Piovesan

Biglan

. Restless legs syndrome and drug-induced akathisia in headache patients. CNS Spectr 2003; 8: 450–456.

34.

Çetīnkaya

Çömez Yılmaz

Türkoğlu

. The relationship between tension-type headache patients with anemia and restless-leg syndrome. J Neurol Sci Turk 2009; 26: 305–310.

35.

Gozubatik-Celik

Benbir

Tan

. The prevalence of migraine in restless legs syndrome. Headache 2014; 54: 872–877.

36.

Stroup

Berlin

Morton

. Meta-analysis of observational studies in epidemiology: A proposal for reporting. Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000; 283: 2008–2012.

37.

Classification and diagnostic criteria for headache disorders cranial neuralgias and facial pain. Headache Classification Committee of the International Headache Society. Cephalalgia 1988: 8 (Suppl 7): 1--96.

38.

Walters

. Toward a better definition of the restless legs syndrome. The International Restless Legs Syndrome Study Group. Mov Disord 1995; 10: 634–642.

39.

Higgins

Thompson

Deeks

. Measuring inconsistency in meta-analyses. BMJ 2003; 327: 557–560.

40.

Galbraith

. A note on graphical presentation of estimated odds ratios from several clinical trials. Stat Med 1988; 7: 889–894.

41.

Begg

Mazumdar

. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994; 50: 1088–1101.

42.

Egger

Davey Smith

Schneider

. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315: 629–634.

43.

Gunalan

Chapman

Champion

. Migraine and non-migraine headaches in children and adolescents: A twin family case-control study of genetic influence, pain and psychological associations. Anaesth Intensive Care 2012; 40: 539.

44.

Akdanull Uzun

Kurt

Karaer Unaldi

. Migraine and restless legs syndrome. Eur J Neurol 2011; 18(Suppl 2): 213.

45.

Wong

Coenders

Findeisen

. Relationships between growing pains, restless legs syndrome, migraine and headaches: A twin family case control study. Anaesth Intensive Care 2011; 39: 754–755.

46.

Coenders

Champion

Hannaford

. Associations between chronic pain disorders in adolescents and history of functional pain syndromes: A case-control study. Anaesth Intensive Care 2011; 39: 741.

47.

Kruit

Launer

Overbosch

. Iron accumulation in deep brain nuclei in migraine: A population-based magnetic resonance imaging study. Cephalalgia 2009; 29: 351–359.

48.

Trenkwalder

Paulus

. Restless legs syndrome: Pathophysiology, clinical presentation and management. Nat Rev Neurol 2010; 6: 337–346.

49.

Silberstein

Young

Mendizabal

. Acute migraine treatment with droperidol: A randomized, double-blind, placebo-controlled trial. Neurology 2003; 60: 315–321.

50.

Bergerot

Storer

Goadsby

. Dopamine inhibits trigeminovascular transmission in the rat. Ann Neurol 2007; 61: 251–262.

51.

Garcia-Borreguero

. Time to REST: Epidemiology and burden. Eur J Neurol 2006; 13(Suppl 3): 15–20.

52.

Park

Lee

Kang

. Prevalence of idiopathic and secondary restless legs syndrome in Korean women. Gen Hosp Psychiatry 2010; 32: 164–168.

53.

Szentkiralyi

Völzke

Hoffmann

. Multimorbidity and the risk of restless legs syndrome in two prospective cohort studies. Neurology 2014. Epub ahead of print 7 May 2014. DOI: 10.1212/WNL.0000000000000470.

54.

The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808.

Migraine and restless legs syndrome: A systematic review

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Subjects and methods

Data sources and searches

Study selection

Data extraction

Statistical analysis

Results

Study characteristics

Prevalence of RLS in migraine/headache

Prevalence of migraine/headache in RLS

Association between migraine/headache and RLS

Discussion

Clinical implications

Footnotes

Funding

Conflict of interest

References