Introduction

Multiple sclerosis (MS) is an autoimmune disease, affecting women more than men with a wide range of complications [1]. Physical and mental complications lead to impaired quality of life and interferes with daily activities [2]. Restless legs syndrome (RLS), extrapyramidal hyperkinesia, is considered as a sleep-related movement disorder according to the German Society for Sleep Research and Sleep Medicine (DGSM) [3]. The International Restless Legs Syndrome Study Group (IRLSSG) revised the diagnostic criteria and declared the characteristics of RLS as strong urge to move limbs along with unpleasant sensations which worsens at rest and in the evening [4].

For the first time, in 2005, RLS in MS cases was introduced, and up to now, various studies reported a wide range of prevalence of RLS in MS cases, ranging from 14 to 65% [5,6,7,8].

The last systematic review and meta-analysis which was conducted in 2018 estimated the pooled prevalence of RLS as 20% which is higher in women [7].

The goal of this systematic review and meta-analysis is to update the prevalence of RLS in MS cases.

Methods

Literature search

We searched PubMed, Scopus, EMBASE, CINAHL, Web of Science, Google Scholar, and gray literature including references from identified studies and conference abstracts which were published up to June 2021.

Inclusion criteria were cross-sectional/case–control studies evaluating prevalence of RLS in patients with MS.

Exclusion criteria were letter to editors, cohort, or randomized clinical trials.

Data search and extraction

The search strategy included the MeSH and text words as ((Sclerosis AND multiple) OR (sclerosis AND disseminated) OR “disseminated sclerosis” OR “multiple sclerosis” OR “acute fulminating”) AND(“Restless Leg*” OR “Willis Ekbom Disease” OR (Disease AND “Willis Ekbom”) OR “Wittmaack-Ekbom Syndrome” OR (Syndrome AND “Wittmaack-Ekbom”) OR “Willis-Ekbom Disease” OR (Disease AND “Willis-Ekbom”) OR “Willis-Ekbom Syndrome” OR (Syndrome AND “Willis-Ekbom”) OR “Wittmaack Ekbom Syndrome” OR (Syndrome AND “Wittmaack Ekbom”) OR “Restless Leg Syndrome” OR (Syndrome AND “Restless Leg”) OR “Willis Ekbom Syndrome” OR (Syndrome AND “Willis Ekbom”)).

Two independent researchers independently assessed the articles. Data on the total number of participants, first author, country, disease duration, number of controls, mean patient age, male and female numbers, mean EDSS, and number of cases and/or controls with RLS were extracted from the included studies.

Risk of bias assessment

We evaluated the risk of potential bias by the Newcastle–Ottawa Quality Assessment Scale (adapted for cross-sectional/case–control studies) [9, 10].

Statistical analysis

All statistical analyses were performed using STATA (version 13.0; Stata Corp LP, College Station, TX, USA).

Inconsistency (I2) was calculated to determine heterogeneity.

Subgroup analysis was done based on sex.

Results

The literature search revealed 855 articles; after deleting duplicates, 530 remained. For the meta-analysis, 75 studies were included (Fig. 1). In six articles, the authors did not differentiate between CIS and MS cases when reporting RLS cases.

Fig. 1
figure 1

Flow diagram of including studies

Full size image

Studies were published between 1999 and 2021. Most studies were conducted in Turkey and the USA. Twenty-two studies reported data regarding the control group. Mean age and mean EDSS ranged between 28 and 59 years, 0.9 and 5.3, respectively.

Totally, 15,411 MS/CIS patients were evaluated and 4309 had RLS. As controls, 66,053 were enrolled and 4496 cases with RLS were reported.

Risk of bias which was assessed by NOS was more than 5 which indicates that the quality of included studies was satisfactory.

The data extracted from studies are summarized in Table 1.

Table 1 Data extracted from the studies
Full size table

The pooled prevalence of RLS was 28% (95% CI: 24–33%) (I2 = 98.3%, P < 0.001) (Fig. 2).

Fig. 2
figure 2

The pooled prevalence of RLS in MS/CIS cases

Full size image

The pooled prevalence of RLS in women was 30% (95% CI: 25–35%) (I2 = 90.6%, P < 0.001) (Fig. 3).

Fig. 3
figure 3

The pooled prevalence of RLS in female patients

Full size image

The pooled prevalence of RLS in men was 22% (95% CI: 17–26%) (I2 = 78%, P < 0.001) (Fig. 4).

Fig. 4
figure 4

The pooled prevalence of RLS in male patients

Full size image

The pooled prevalence of RLS in controls was 8% (95% CI: 6–10%) (I2 = 99.5%, P < 0.001) (Fig. 5).

Fig. 5
figure 5

The pooled prevalence in controls

Full size image

The pooled odds of RLS in patients with MS was 4.03 (95% CI: 1.83–3.57) (I2 = 64.8, P = 0.001) (Fig. 6).

Fig. 6
figure 6

The pooled OR of developing RLS in the MS group compared with controls

Full size image

Discussion

This systematic review and meta-analysis is the update of previous studies, including 75 studies.

We found that the pooled prevalence of RLS in subjects with MS was estimated as 28%, while the pooled prevalence in controls was estimated as 8%.

The pooled prevalence of RLS was higher in women than men (30% vs 22%).

The previous systematic review and meta-analysis was conducted by Ning et al. They included 25 articles and reported the pooled prevalence of RLS in patients with MS as 27%. In their study, like our study, the pooled prevalence was higher in females than males (26 vs 17%). They also found that the odds of RLS was near fourfold among MS cases than controls [7] which is in agreement with our findings (OR = 4.03).

In another systematic review which was conducted in 2013 by Schürks, they investigated that the prevalence of RLS in subjects with MS ranged between 12.1 and 57% and also the prevalence ranged between 2.5 and 18.3% in controls. They reported the pooled OR of 4.1 (95% CI: 3.1–5.6) for RLS in MS [84]. They did not report the pooled prevalence.

Bruno et al. enrolled 152 patients with MS and 431 healthy controls and reported RLS as 14.5% in MS group vs 6% in controls. They found that the presence of cervical cord lesions was associated with RLS development (OR = 3.7, 95% CI: 1.1–13.5) [85].

Liu et al. evaluated 695 individuals with MS and 603 controls and found the prevalence of RLS as 24.6% in MS group vs 8% in controls. They noticed that RLS in MS group was more severe than controls and sleep quality in MS patients with RLS was more impaired [50].

In a recent study which is conducted by Monschein et al., the prevalence of RLS estimated as 23.9% in MS group and 3.4% in healthy group and MS cases with RLS had higher proportion of sleep impairment and excessive daytime sleepiness. They also found that disability level and spinal lesions in MRI are independent predictors of RLS development in MS [12]. Maybe spinal lesions disturb hypothalamo-spinal dopaminergic system [86], leading to RLS in MS.

Minar et al. enrolled 200 MS cases and reported RLS in 26% while RLS development was assumed to be associated with spinal cord lesions (OR = 3.846, 95% CI: 1.304–11.346) [87].

RLS is a sensory motor disorder which characterized by an impulse to move the legs that may be accompanied by dysesthesias and unpleasant sensation [88]. Untreated RLS is associated with sleep disturbances, daytime sleepiness, depression, and impaired quality of life [88]. The average sleep time in cases with RLS estimated as 5 h a day [88].

The diversity between studies regarding the prevalence of RLS in patients with MS could be due to variety of diagnostic method. In some studies, International RLS Rating Scale (IRLSS)/ Cambridge-Hopkins Questionnaire (CH-RLSq) and others administered clinical interview. We did not do meta-analysis based on diagnostic method as some studies which are included in our study did not report the diagnostic report.

In general population, iron deficiency is related with presence of RLS, while iron deficiency and vitamin D and B deficiencies were not associated with RLS in MS patients [87].

As reported previously, disability level and location of MS-related lesions were associated with RLS development.It is shown that infra-tentorial lesion increase the risk of RLS by sevenfold [89].

Administration of disease-modifying therapies have various effects on RLS development [89]. Some researchers found that interferons negatively affect sleep quality and daytime sleepiness leading to RLS worsening in subjects with MS [16, 90, 91]. On the other hand, natalizumab use is related with sleep quality improvement [92]. The controversies are based on cross-sectional nature of studies.

This systematic review has some strengths. First, we calculated the pooled prevalence in patients and controls. Second, we included 75 articles. Third, the pooled prevalence is calculated in both sexes separately.

Conclusion

The results of this systematic review and meta-analysis show that the pooled prevalence of RLS is 28% in MS cases and 8%. The pooled prevalence is higher in women than men (30% vs 22%).