European Journal of Neurology 2007, 14: 1275–1280
doi:10.1111/j.1468-1331.2007.01966.x
Low RLS prevalence and awareness in central Greece: an epidemiological survey G. M. Hadjigeorgioua,b, I. Stefanidisb,c, E. Dardiotisa,b, K. Aggellakisa,b, G. K. Sakkasc, G. Xiromerisioua,b, S. Konitsiotisd, K. Paterakise, A. Poultsidif, V. Tsimourtoua, S. Rallia, K. Gourgoulianisg and E. Zintzarash a
Department of Neurology, Medical School of Larissa, University of Thessaly, Larissa, Greece; bBiomedical Institute of Research and Technology, Larissa, Greece; cDepartment of Nephrology, Medical School of Larissa, University of Thessaly, Larissa, Greece; dDepartment of Neurology, Medical School of Ioannina, University of Ioannina, Ioannina, Greece; and Departments of eNeurosurgery and fSurgery, g
Respiratory Medicine and hBiomathematics, Medical School of Larissa, University of Thessaly, Larissa, Greece
Keywords:
epidemiology, Greece, prevalence, restless legs syndrome Received 16 April 2007 Accepted 3 September 2007
Restless legs syndrome (RLS) is a sensorimotor disorder with a general population prevalence of 3–10%. A single, previous epidemiological study performed in southeast Europe reported the lowest prevalence rate amongst European countries. We conducted a population-based survey of RLS in central Greece. A total of 4200 subjects were randomly recruited. We used the international RLS study group criteria for diagnosis and the severity scale for severity assessment in subjects with RLS. We also included questions to assess the level of awareness of RLS in our region. A total of 3033 subjects were screened. The overall lifetime prevalence was 3.9% with a female-to-male ratio of 2.6:1. Nearly half of RLS patients reported moderate to severe intensity of symptoms. After adjustment for multiple comparisons we found no association of RLS with education level, smoking, alcohol intake, caffeine consumption, shift work, professional pesticide use or comorbid illness. Our study revealed a low level of awareness amongst the population and physicians in our region and suboptimal management. We provide further evidence for low prevalence of RLS in south-east Europe and a low level of awareness of RLS in our region.
Introduction Restless legs syndrome (RLS), first extensively described by Ekbom in 1945, is a common movement disorder characterized by unpleasant and uncomfortable leg sensations and an urge to move the legs [1–3]. The exact occurrence of RLS in the general population is still unknown, but it is probably that most cases remain undiagnosed [3–7]. Published, revised clinical diagnostic criteria from the International RLS Study Group (IRLSSG) have recently been used in many epidemiological studies to better define the frequency of RLS [8]. If only IRLSSG criteria-based studies are considered, epidemiological studies reveal a striking ethnic variation and an increased prevalence of RLS with age [3,9]. Women are affected more than men in all age categories except for subjects aged ‡80 years, in whom RLS symptoms have a higher prevalence in men than women [3,10]. In western countries the RLS Correspondence: Georgios M. Hadjigeorgiou, Department of Neurology, Medical School, University of Thessaly, Papakyriazi 22 Street, Larissa 41222, Greece (tel.: +30 697 2862 909; fax: +30 241 0681 667; e-mail:
[email protected]).
2007 EFNS
prevalence rate ranges from 3.2 to 11.5% in the general population, whilst in Asian countries population studies indicate a prevalence rate 2 drinks/day versus £2 drinks/day for the last month), caffeine consumption (>2 cups/day versus £2 cups/day for the last month), shift work (during last year, yes versus no), professional use of pesticides (during last year, yes versus no) and (ii) medical history for co-morbid illness (based on current treatment), hypertension, diabetes mellitus, chronic kidney disease and anaemia. To explore awareness of RLS, we raised the following question for both RLS and control group: have you ever heard of RLS? Subsequently, each participant who fulfilled the four clinical criteria for RLS was asked to visit the University Hospital of Larissa for detailed neurological examination (an option for a visit at home was also offered). The Greek version of the IRLS rating scale was also recorded by a single RLS expert (GMH) for all RLS subjects [13]. Severity was classified as mild (5 times per month) [4]. To clarify the reasons for the possible mis- or underdiagnosis, we also asked participants the following questions: 1) How many years have you had RLS symptoms? We separated them as having symptoms for £10, 11–20, 21–30, 31–40 and >40 years. 2) (i) Have you ever informed your family doctor about your symptoms? (ii) If yes, how did your doctor explain your symptoms? 3) Has any doctor asked you specifically about symptoms resembling RLS symptoms? Patients were then asked to check if RLS symptoms had been attributed to one or more of the following disorders (in their opinion or their physician’s diagnosis): lumbar spine disorder, depression/anxiety, problem with varicose veins or perfusion problem of the legs, polyneuropathy and osteoporosis. Statistical analysis
Randomization was performed with stratified systematic random sampling from the local registry [15]. Differences between mean values of continuous variables (age and years of education) were tested using the t-test, whilst proportions for dichotomous variables were calculated with chi-square and Fisher exact tests, as appropriate. For dichotomous variables in Table 1, a two-sided P-value of 10 years. Our data is similar to a cohort study from South Africa recruited from responses to print and radio advertizements [20]. The frequency of smokers in our study population is similar to recent epidemiological studies that estimated that 40% of the Greek adult population smokes daily [21]. Although the frequency of smokers in the RLS group is increased compared with controls, this difference did not reach statistical significance after adjustment for multiple comparisons. The possible association of cigarette smoking with increased risk of RLS has also been investigated in previous studies with inconsistent results [4,7,12,22]. A possible explanation for this discrepancy is the different measurements used to determine smoking habits. In some studies smoking was determined by a single, dichotomous question (yes or no), whilst in other studies investigators were able to separate heaver from lighter smokers. It is possible that increased risk for RLS is true only for heavier smokers. Adverse health effects after chronic occupational low-level exposure to organophosphates have recently been elucidated [23]. Some studies suggest that longterm effects on the central and peripheral nervous system may be associated with low-level exposure to organophosphates. To our knowledge, this is the first study to investigate the possible role of pesticide use in RLS. We failed to identify any increased risk for RLS in subjects with a history of professional use of pesticides during the last year. Recent reports suggest a clear circadian rhythm for subjective complaints of RLS and changes in melatonin secretion were found to precede the increase in sensory and motor symptoms [2,24]. In the present study, we investigated whether shift work during the last year was associated with RLS symptoms. Although the percentage of shift workers was higher in the RLS group compared with controls, this difference was not significant. As our study is observational, it is possible that the prevalence of comorbidity is underestimated, albeit in accord with recent reports from Greece [25,26]. Of the known secondary causes of RLS, anaemia and chronic kidney disease showed increased frequency, but not significantly so, in the RLS patient group compared with controls after applying correction for multiple comparisons. Several papers have pointed to a strong association between the occurrence of RLS and reduced iron stores, with or without anaemia [3]. Our results, however, do not allow us to infer true causality because diagnosis was based on current treatment status and was not verified with blood tests. The percentage of subjects under treatment for hypertension was similar in the RLS control group. Previous studies report
2007 EFNS European Journal of Neurology 14, 1275–1280
RLS prevalence in Greece
controversial results associating RLS with hypertension [4,7,27–29]. Recently, two studies reported that RLS patients present a periodic (or transient) significant increase in blood pressure during episodes of periodic limb movements in sleep and wakefulness [30,31]. This transient variability may have an adverse effect on the cardiovascular system. The data presented in these studies provide further evidence for the hypothesis that RLS is strongly associated with cardiovascular disease as reported in previous epidemiologic studies [27,28]. The strengths of our study, amongst others, include the use of validated, international RLS criteria, proper randomization, trained interviewers and verification of diagnosis by an RLS expert. It was striking that only a small percentage (
