Introduction

Restless legs syndrome (RLS) is a common sleep disorder in general population with the prevalence of 5–15 %. The symptoms present with a disturbing sensation associated with an irresistible urge to move the legs, especially at evening or at night, which can improve with leg movements. This syndrome can be idiopathic or secondary to pathologic conditions such as pregnancy, iron deficiency anemia, neurologic diseases, and renal insufficiency [1–4].

In hemodialysis patients, the prevalence of RLS has been reported between 12 and 62 % [3], and the associated factors are introduced as iron deficiency, vitamin deficiencies, and diabetes mellitus [3]. Furthermore, there are significant associations between RLS and frequency of dialysis per month, serum levels of blood urea nitrogen (BUN) and creatinine in patients with chronic renal failure (CRF) [5]. Renal transplantation significantly improves RLS symptoms, which further confirms the role of renal insufficiency [6].

RLS causes sleep deprivation, daytime sleepiness, depression, reduced quality of life and increased morbidity in patients with renal failure [3]. RLS increases the cardiovascular risk and mortality in chronic hemodialysis patients [7]. Studies have shown that treatment of sleep disorders in CRF patients can improve the quality of life [4]. The present study aimed to evaluate the frequency and associated factors of RLS in CRF patients treated with hemodialysis.

Methods

This cross-sectional study was conducted in hemodialysis center of Rasool-Akram hospital affiliated to Iran University of medical sciences, during May 2012 to September 2012. The study protocol was approved by the institutional medical ethics committee.

One hundred and sixty-three CRF patients treated with hemodialysis, consecutively enrolled in the study. Inclusion criteria for the study included: being able and willing to give informed consent; being diagnosed with CRF and treated with hemodialysis at least for 1 year; having dialysis at least two times a week. Patients underwent neurological examinations by one neurologist to exclude other secondary neurological causes of RLS. Patients with abnormal neurological exam, including sensory or motor impairments were excluded from the study. Presence of severe infection or acute systemic diseases was other exclusionary criteria. A face to face interview was performed and demographics including age, gender, marital status, education, height, weight and BMI, history of smoking, alcohol and opium consumption were recorded. Patients were asked about duration of night sleep and sleep problems including insomnia, daytime sleepiness, sleep latency, and consumption of sedative-hypnotics.

The type of renal failure, frequency of dialysis per week, duration of dialysis, dialysis dose, history of kidney transplant, diabetes mellitus, and hypertension were recorded from medical records and interview. Laboratory evaluations included hemoglobin level, ferritin, fasting blood sugar, BUN, and creatinine.

Patients were assessed for the presence of RLS by the four criteria of international RLS study group (IRLSSG) during the face to face interview. These criteria include: unpleasant sensations in foots or legs which develop or exacerbate during rest; aggravate in the evening or at night; and relieve with movement. Patients who fulfilled all four criteria considered to have RLS and further evaluated by the self-questionnaire of IRLSSG severity scale. Family history of RLS was also evaluated in RLS-positive patients.

The data were expressed as mean ± standard deviation (SD). Chi square test and t test were used for qualitative and quantitative data, respectively. All statistical analyses were performed with SPSS version 16 for Windows (SPSS Inc. Chicago, IL, USA).

Result

A total of 163 hemodialysis treated CRF patients with mean age of 61.3 ± 13.3 years (range 18–82 years) (63.2 % male) were studied. One hundred and eight patients (66.3 %) had complaint of insomnia more than 2 nights per week, and 71 patients (43/6 %) were experiencing daytime sleepiness more than 2 days per week. Average sleep duration at night was 6.2 ± 1.4 h (range 3–10 h). Sixty-seven patients (41.4 %) were taking sedative-hypnotic medications.

Based on the four IRLSSG criteria, 61 patients (37.4 %) were diagnosed with RLS. Mean age of RLS patients was significantly higher than patients without RLS (65.2 ± 9.3 years vs. 59.0 ± 14.7; P = 0.004). There was no significant association between RLS and gender, marital status, education, drug usage, drug abuse, alcohol intake, smoking, and BMI. Unemployment was related to presence of RLS (P = 0.03) (Table 1).

Table 1 Demographic features of CRF-Hemodialysis patients with and without RLS
Full size table

There was no significant difference between RLS-positive and RLS-negative patients in terms of dialysis frequency and duration, history of renal transplantation, diabetes mellitus and hypertension (P > 0.05). However, the prevalence of cardiovascular diseases was significantly higher in RLS-positive patients (32.8 vs. 16.7 %; P = 0.01) (Table 2).

Table 2 Characteristics of renal failure, dialysis and laboratory tests in CRF patients with and without RLS
Full size table

No difference was observed between the two groups regarding the hemoglobin level, serum level of ferritin, fasting blood glucose and BUN. However, serum creatinine level in patients with RLS was significantly higher than other patients (7.6 ± 2.1 vs. 6.7 ± 1.8 mg/dl) (P = 0.009). Furthermore, glomerular filtration rate (GFR) in RLS-positive patients was significantly lower (9.2 ± 3.1 vs. 11.6 ± 4.8 ccs/min; P = 0.0001) (Table 2).

Sleep complaints were more common in RLS-positive patients (P = 0.0001). These complaints included shorter duration of sleep, more sleep latency, frequent experience of insomnia and daytime sleepiness, and more common sedative-hypnotic consumption (P = 0.0001) (Table 3).

Table 3 Sleep complaints in RLS-positive and RLS-negative patients
Full size table

The average RLS severity score was 22.7 ± 7.2 (Max: 40). Three patients (4.9 %) had mild symptoms (score 0–10), 23 patients (37.7 %) had moderate symptoms (score of 11–20), 22 patients (36.1 %) reported severe symptoms (score of 21–30), and 13 patients (21.3 %) had very severe symptoms (score of 31–40). There was no significant association between severity score and age, gender, BMI, serum levels of glucose, blood urea nitrogen, creatinine, ferritin, and GFR (P > 0.05). There was an inverse correlation between severity score and hemoglobin levels (P = 0.009, r = −0.33). Positive family history of RLS was reported in 6 patients (9.8 %) and it was not related to age, gender, RLS duration and severity, or laboratory findings (P > 0.05).

Discussion

Prevalence of restless legs syndrome in patients with CRF (19–56.4 %) [8–16], especially those treated with hemodialysis (12 and 62 %) [3], has been reported significantly higher than general population (5–15 %) [1–4]. The prevalence of RLS in our hemodialysis patients was 37.4 %. We observed no association between RLS and gender, marital status, education, drug usage, alcohol intake and smoking in hemodialysis treated patients. Salman and colleagues have also reported no association between RLS incidence and sex, nicotine and caffeine consumption [10]. Demographics, smoking, caffeine intake, and higher BMI are recognized as risk factors for RLS in healthy population [1, 2]. Our findings, consistent with previous reports, suggest that demographics features have a lower impact on these patients’ population. We only observed the relationship between RLS and higher age in our CRF patients. Pizza et al. (2012) [17] reported a higher rate of positive family history of RLS (22 %) in hemodialysis patients with RLS compared to those without RLS (6 %). Among our RLS-positive patients, 9.8 % had a positive family history of RLS.

Peripheral neuropathy and first-degree relative with RLS are predictive factors of RLS in hemodialysis patients [17, 18]. Peripheral neuropathy in CRF patients can cause moderate severity of leg discomfort which may be misdiagnosed with RLS. Careful neurological exam along with IRLSSG criteria is suggested for RLS evaluation in CRF patients [18].

This study found no association between incidence of RLS and type of renal failure, dialysis frequency, dose and duration of dialysis, history of renal transplant, diabetes and hypertension. Similarly, Nikic and colleagues reported no differences between duration of dialysis and features of hemodialysis in patients with and without RLS [8]. Current study indicates that previous history of ischemic heart disease is associated with RLS. Zilberman and colleagues have shown the high prevalence of RLS in the anemic patients who had chronic renal failure or congestive heart disease (39.4 %) [19]. The finding can be due to negative impact of cardiac failure on renal function which may augment the underlying cause of RLS in CRF patients.

Previous studies have concluded that lower GFR (15–29 ml/min) is a risk factor for RLS [20]. Pizza et al. introduced low or absent residual diuresis as predictive factors of RLS in hemodialysis patients [17]. We observed the association RLS with higher serum creatinine level and lower GFR. Similarly, the study of Nikic and colleagues reported that creatinine level in patients with RLS was higher than other patients, which could be indicative of inadequate dialysis in these patients [8]. Study of Ibrahim and his colleagues showed that sleep disorders and the incidence of RLS are significantly related to inadequate dialysis, hyperphosphatemia, and hypoalbuminemia [9].

The high prevalence of RLS in end stage renal disease patients undergoing hemodialysis can be extremely troublesome. Sleep disorders in hemodialysis patients have been reported from 46 to 61 % [9, 13, 14]. The study by Merlin and colleagues has shown that approximately 80 % of hemodialysis patients have at least one sleep disorder [16]. Prevalence of insomnia and other sleep disturbances in current study was significantly higher than previous studies and these disturbances were more common in RLS-positive patients (88 %). Shorter sleep duration, insomnia, sedative drugs usage, and daytime sleepiness were more common in patients with RLS. Tuncel’s study discussed about the quality of Epworth sleepiness scale (ESS) showing that 47 % of hemodialysis patients, especially those with RLS, have sleep disorders [14]. Araujo and colleagues concluded that lower hemoglobin levels in hemodialysis patients are associated with more severe RLS, worse sleep quality, higher scores on the Beck Depression Inventory, more daytime sleepiness and obstructive sleep apnea [15]. Previous studies have shown that undiagnosed RLS can significantly reduce quality of life in CRF patients [11, 19]. Therefore, accurate diagnosis and appropriate treatment of RLS can improve quality of life in these patients.

In conclusion, RLS has a high prevalence in CRF patients treated with hemodialysis. Presence of RLS in hemodialysis patients is related to higher levels of serum creatinine and lower GFR which suggests the direct relationship between renal dysfunction and RLS. Furthermore, this study shows poor sleep as a consequent of RLS in hemodialysis patients which indicates the importance of RLS screening and treatment in these patients population.