Introduction

Restless legs syndrome (RLS) is a sensorimotor-related sleep disorder that is commonly under-diagnosed, misdiagnosed, and poorly understood [1–5]. Unfortunately many people suffer from RLS, and only 6.2% of RLS patients received the appropriate RLS diagnosis and management [2, 6]. The International Restless Legs Syndrome Study Group (IRLSSG)defines RLS by the presence of four criteria: (1) an urge to move the limbs with or without sensations, (2) worsening while resting, (3) improvement with activity, and (4) worsening in the evening or night [7].

The true prevalence of RLS is variable and has been ranging from 1 to 29% [8–14]. One local study reported a prevalence of 5% among patients visiting primary care professionals [15] and another study reported a prevalence of 8.4% among middle-age school employees [16]. RLS can affect 5 people of any age, but mostly affects those in middle age. It increases with age, and while it can affect some children, and more women, a few other studies reported more prevalence in males [12, 17–25]. The underlying etiologies of RLS are many. RLS is associated with many conditions including smoking, low socioeconomic status, obesity, metabolic and hormonal changes, pregnancy, iron deficiency, anemia, renal failure, and neuropathy [26–32]. Family history is present in 40–50% of the cases, particularly in the young population [33, 34]. RLS is usually associated with significant sleep disturbance [29, 35–39]. Most of the patients may present with complaints related to the consequence of the disease, such as sleep disturbance, and may include impairment of sleep onset, sleep maintenance, impaired daytime functions, symptoms of insomnia, depression, anxiety, impaired cognition and memory, poor sleep quality or excessive daytime sleepiness (EDS), generalized fatigability, and poor health-related quality of life [12, 29, 35–47]. The clinical course of RLS is variable from mild to severe, but the symptoms tend to wax and wane overtime. Unfortunately, RLS is commonly under-recognized and under-diagnosed by physicians, and patients may have difficulties in describing their symptoms, which may also contribute to delaying the diagnosis [20]. In this study, we aimed to examine the prevalence of RLS and the relationship between RLS and other sleep disorders, quality of sleep, excessive daytime sleepiness, and associated health conditions among a sample of the healthy adult Saudi population.

Methodology

This study is an observational cross-sectional study and was conducted among samples of Saudi people at King Abdul-Aziz Medical City—Riyadh (KAMC). King Abdulaziz Medical City has been recognized as a major healthcare provider in Kingdom of Saudi Arabia (KSA). The attendees of King Abdulaziz Medical City represent a spectrum of different backgrounds and socio-demographic characteristics. We chose attendees of King Abdulaziz Medical City in Riyadh as the population in this study. The study participants were recruited from pre-employment clinic, blood donors, visitor to KAMC, and employees.

Trained interviewers were assigned to obtain written informed consent and to ensure that all questionnaires were answered by the participants. Data were collected about demographic characteristics such as age, gender, education level, and marital status, employment, past medical history, medication, smoking, and coffee intake. The data collection was carried out using the Arabic version of structured survey questionnaires which uses the International Restless Legs Syndrome Study Group (IRLSSG) for the clinical diagnosis of RLS [7]. We assessed EDS using the Epworth Sleepiness Scale (ESS) with a score >10 indicating EDS [48] and the Berlin Questionnaires to estimate risk for sleep apnea [49]. Pittsburg Sleep Quality Index (PSQI) was used to assess sleep quality (a score of more than 5 indicates good sleep quality) [50]. All these questionnaires have been validated for the Arabic language and were used in previous studies [15, 16, 51–53].

Statistical methods

All required statistical analyses were carried out using IBM Statistical Package for Social Sciences (SPSS®) version, 22.0 (Chicago, IL, USA).Data were presented in tables and graphs. Data were presented as means and standard deviations for continuous variables and as percentages for categorical variables. Comparisons between groups to assess the possible influence of demographic data and other variables on the prevalence of RLS were performed using the unpaired T test, or the Mann–Whitney U test for nonparametric data as appropriate. Age and gender effect on the presence of RLS was analyzed by the Mantel–Haenszel statistical test. A multivariable logistic regression analysis was used to identify significant independent risk factors for RLS while controlling for other parameters, including, age, gender, comorbidities, ESS (a score higher than 10 indicated EDS), risk for sleep apnea (using the Berlin Questionnaire, high risk vs. low risk), and PSQI (a score higher than 5 indicated poor sleep quality). We also investigated the severity of RLS in respondents who met the criteria of RLS using International Restless Legs Syndrome Study Group rating scale for restless legs syndrome for severity assessment [54]. The strength of the associations was assessed by odds ratio and 95% confidence intervals. A P value of less than 0.05 will be considered statistically significant.

Results

From a total of 2250 questionnaires distributed, only 2095 surveys were completed, giving a 93% response rate. Of 2095, 59.3% were males, 71.1% were elderly (aged 60 years or more), 20.5% were smokers, and 39.1% were obese. In our sample, the prevalence of RLS was 14.7% with 95% confidence intervals (CI) of 13.3–16.3%. The prevalence of RLS and related demographic and sleep characteristics are shown in Table 1. The presence of RLS was higher in males than females (16.3% in males vs. 12.5% in females, P = 0.016). The presence of RLS was higher among smokers than non-smokers (19.5 vs. 13.5%, P = 0.002). RLS was significantly more common in respondents with depression (27.5 vs. 14.2%, P = 0.001), short sleep duration less than 6 h (21.7 vs. 14.1%, P = 0.005), hyperlipidemia (29.5 vs. 15.7%, P = 0.013), poor sleep quality (16.2 vs. 9.7%, P = 0.001), and high risk for sleep apnea (19.0 vs. 12.8%, P = 0.001). The prevalence of RLS was not affected by various age groups (14.1% in the young age group (18–29), 15.9% in the middle-age group (30–60), 11.7% in the elderly (over 60), P = 0.139). However, an age difference was found when we stratified the prevalence of RLS by gender. The RLS significantly decreased among elderly females more than young or middle-aged female groups (8.3% in elderly females vs. 14.9% in middle-aged females and 11.4% in young females, P value = 0.049). Age and gender effects on the presence of RLS are shown in Fig. 1.The association with educational level, marital status, or employment status was not significant (P value >0.05).

Table 1 Prevalence of RLS and related demographic and sleep characteristics
Full size table
Fig. 1
figure 1

Age and gender effects on the presence of RLS

Full size image

We further analyzed the severity of RLS and related factors in respondents with the presence of RLS (N = 309). Among participants with RLS, 20.8% reported severe RLS, 30.7% reported moderate RLS, and 48.5% reported mild RLS. The severity of RLS decreased with age (72.2% in the young group, 50% in middle-age group, and 14.6% in the elderly, P = 0.001). The severity of RLS was significantly higher in smokers compared with non-smokers (67.5 vs. 45.3%, P = 0.001). The severity of RLS was significantly related to depression (92 vs. 47.9%, P = 0.001), short sleep duration (80.5 vs. 47.0%, P = 0.001), and poor sleep quality (55.8 vs. 25.0%, P = 0.001). The effect of demographic and sleep characteristics on the severity of RLS are shown in Table 2. The severity of RLS was 15 times higher in the young group (adjusted Odds Ratio [aOR] = 15.52; 95% CI 3.668–65.627; P = 0.001) and 4.88 times higher in the middle-age group (aOR = 4.88; 95% CI 1.363–17.484; P = 0.015), compared to the elderly. The RLS was 40 times more severe in respondents with depression (aOR = 40.21; 95% CI 1.199–2.372; P = 0.003), Table 3.

Table 2 RLS severity demographic and sleep characteristics
Full size table
Table 3 Risk factors of RLS severity
Full size table

Risk factors of RLS using the multivariate logistic regression model are shown in Table 4.The model revealed that smoking (aOR = 1.686; 95% CI 1.199–2.372; P = 0.003), short sleep duration (aOR = 1.568; 95% CI 1.024–2.400; P = 0.039), poor sleep quality (aOR = 1.969; 95% CI 1.236–3.139; P = 0.004), and high risk of sleep apnea (aOR = 1.598; 95% CI 1.113–2.295; P = 0.011) were significantly associated with the presence of RLS.

Table 4 Risk factors of RLS using multivariate logistic regression
Full size table

Discussion

RLS is reported as a common important health problem in Saudi Arabia [16], and is associated with significant poor quality of life. The prevalence of RLS found in our study was 14.6%, which is comparable with that in the Caucasian populations of Europe and North America, and is much higher than other local studies [15, 16]. The RLS prevalence was reported with a wide variation: 2–16% in the Indian population [5, 55], 6% in the Turkish population [56], 1–11% in the Chinese and Japanese populations [57–60], 8.3–12% in the Korean population [23, 61, 62], 0.6% among general population in Singapore [63] and 20–45% in Caucasians [13, 64, 65].This suggests that genetic differences may play a role. This discrepancy may be attributed, at least partially, to the differences in the sampling methods. Although the prevalence reported in our study is in concordance with the prevalence of values reported by European and North American studies [13, 64, 65], some of our findings were not consistent with those studies [5, 55]. Studies from North America and Europe demonstrated a linear trend of an increase in age with increasing RLS frequencies, but this observation was not found in our study and the studies from Asia [21].

Furthermore, unlike most studies, our study showed that more males than females were affected by RLS. One study recently published from Saudi Arabia shows the same finding [16]. Additionally, it is consistent with a Japanese study that reported a male predominance in individuals aged 40–49 years [7]. Most studies have reported the prevalence of RLS to be approximately twice as high in women as in men. Another important local study reveals that the prevalence of RLS among Saudis was 5.2%, and was highest among participants between 45 and 60 years of age, with an overall prevalence that was equal between males and females [15].

RLS is either idiopathic in nature or occurs secondary to many risk factors. In this study, RLS was not significantly associated with DM, asthma, or COPD. However, we found a significant association between obesity and RLS. This is similar to findings in previous epidemiological studies that indicated that body mass index (BMI), independent of all other confounders, was associated with a higher risk of RLS [66].This relationship may be due to a decrease in the number of dopamine receptors in the brains of obese subjects. Similar to other studies, we also found smoking as a risk factor for RLS [67]. Unfortunately, we did not examine for consanguinity (a marriage between first-degree cousins) which is common in our society, however, one local study documented a significant association with consanguinity and RLS in our population [16].

This study similar to many other studies, demonstrated that RLS was significantly associated with insomnia, poor sleep quality, short sleep duration, excessive EDS and high risk for sleep apnea [8, 29, 36, 39, 68–70]. The painful sensations of RLS is the primary cause of difficulties initiating sleep, insomnia, and poor sleep quality, anxiety, depression and EDS [8, 29, 36, 39, 68]. RLS is also commonly seen in patients with OSA, but there are very few studies that examined the prevalence of RLS in this population [71]. One study documented 8.3% prevalence of clinically significant RLS in OSA patients compared with 2.5% in the control group [72]. However, there was no direct relationship between the severity of the RLS or and OSA severity [70]. The precise mechanism involved in the pathogenesis of RLS among OSA patients is unknown. Few reports link effect of intermittent hypoxia due to apnoeic events on the dopaminergic pathway dysfunction and brainstem as possible relationship between OSA and RLS [70, 73, 74].

Although depressive symptoms have been reported frequently in patients with RLS, the relationship between RLS and depression is not clear [8, 68, 69]. The high association of RLS with depression could be explained in that several factors such as RLS could cause depression through nocturnal sleep disturbances and poor quality of life. This was supported by our observation that RLS patients showed poor sleep quality as measured by PSQI scores, fatigue, diminished concentration, and insomnia as depressive symptoms. We believe that the advantage of this study, compared to the two previous studies done among the Saudi population [15, 16], is that it is a population-based study with a large sample size, and includes all age spectrums among healthy individuals. It is different from a previous study, which was only among primary health care visitors [15] or among middle-age school teachers [16]. The limitations of this study include not examining all other risk factors, consanguinity, the limitation of the criteria for diagnosis of RLS which is based on questionnaires with no gold standard for the diagnosis, and also the fact that the prevalence rate likely encompasses both primary and secondary RLS. Other limitations include no details of information about medications used by patients. In addition, we did not examine for secondary causes of RLS, such as iron deficiency anemia or neuropathy among participants, nevertheless, this is a survey questionnaire where such limitations are unavoidable. The findings of the study may only represent KAMC attendees as the sample recruited may not represent the general population. In addition, cases that mimic RLS may have also been included, leading to an exaggerated prevalence rate.

Conclusion

RLS is common among our population, more so among males than females. Few risk factors were identified. Further studies to define the burden of the disease and the causality relation with the associated conditions are needed in our area.