Prevalence of restless legs syndrome and associated factors in rural West African country

INTRODUCTION

Restless legs syndrome (RLS) is characterized by uncomfortable sensations in the legs, particularly during sleep or rest, causing an urge to move the legs.^[1] In subSaharan Africa, RLS is often under-recognized and underestimated by non-neurologists. This chronic sensory-motor disorder was first described by Ekbom and Ulfberg.^[1] The prevalence of RLS ranges from 0.013% to 23% with most studies conducted on Caucasian populations, showing higher rates in North America and Western Europe.^[2-4] In Africa, few studies have investigated the prevalence of RLS in rural areas or even in hospital settings.^[3,5]

The pathophysiology of primary RLS remains unclear. Three main contributing factors have been proposed although they are not independent^[6] dysfunction of the nigrostriatal dopaminergic system, iron and ferritin, and genetic predisposition.^[7] Women are affected twice as often as men. Risk factors for RLS include female gender, pregnancy, lower socioeconomic status, poor health, low iron levels, advanced age, comorbidities such as Parkinson’s disease or psychiatric disorders, and family history of similar conditions.^[6,7] Given the impact of RLS on quality of life and its potential cardiovascular risk, studying the conduct, prevalence, and associated factors of RLS in rural areas of Benin is warranted.

MATERIALS AND METHODS

RESULTS

Although the survey was initially planned for 1294 participants, 1655 individuals were ultimately enrolled and completed the IRLSSG questionnaire. The mean age was 41.44 ± 16.71 years, with ages ranging from 18 to 93 years. The sex ratio was 0.5, with uneducated individuals comprising 53.5% of the sample. The most common occupations were craftsmen (38.7%), retailers (23.1%), and farmers (21.4%). Among participants, 18 were diabetic and 101 reported neuropathic pain. The mean body mass index (BMI) was 23.7 ± 4.8 ranging from 13.0 to 49.8 and abdominal circumferences ranged from 56 to 151 cm, with an average of 82.1 ± 10.9 cm. HBP was diagnosed in 728 participants (44%; 95% CI: 41.6–46.4%) and Table 1 provides a summary of participants characteristics. HIV rapid diagnostic tests and glycemia measurements were performed on 1339 (80.9%) and 1088 (65.7%) of participants, respectively. The frequency of HIV was 1.2% (16 participants), and 6.1% (66 participants) had hyperglycemia (>1.26 g/L).

The prevalence of RLS was 0.97% (95% CI: 0.50–1.44%) with 16 cases out of 1655 participants, including 7 moderate cases, 5 severe cases, 3 mild cases, and 1 very severe case. Of those with RLS, 56.25% were under the age of 50. There were 9 men and 7 women with RLS, resulting in a sex ratio of 1.29. Among RLS cases, 56.25% had a BMI ≥25, and 62.5% had abdominal obesity.

In the bivariate analysis, RLS was significantly associated with abdominal obesity (OR = 2.9, 95% CI: 1.1–8.2, P = 0.029) and obesity (BMI ≥25) (OR = 2.9, 95% CI 1.1–7.8, P = 0.030), as shown in Table 2. In the multivariate analysis, after adjusting for other characteristics, abdominal obesity remained significantly associated with RLS (ORa = 2.9, 95% CI 1.1– 8.1). Table 3 provides a summary of the multivariate analysis.

DISCUSSION

The prevalence of RLS in our study was 0.97%, based on the IRLSSG criteria. Ekbom and Ulfberg estimated prevalence rates between 1 and 5%.^[1,4] Since 1995, several epidemiological studies using the IRSSLG criteria have reported higher prevalence rates of 5–15% in population studies and predominantly conducted in Caucasian populations, with noted variations across geographic regions.^[4]

In Africa, few studies have examined RLS prevalence. The first published study by Winkler et al., in a rural population in northern Tanzania, included 7654 individuals and found a low prevalence of 0.013%.^[3] More recently, a study in Tanzania’s urban area (Dar es Salaam, Kinondoni) involving a sample of 35,008 participants reported a prevalence of 0.037%,^[5] using the IRLSSG criteria with an adaptation of the first question.

In Europe, RLS prevalence ranges from 3.19% to 25%.^[8-14] In the United States, several population studies using a single-question telephone survey found a prevalence of approximately 9.4%.^[15] Asian populations show a lower prevalence, from 0.1% to −0.6%.^[16-21] However, prevalence rates in South Korea vary significantly due to methodological differences. In 2008, it was based on IRLSSG criteria, compared to 0.9% in 2009 using diagnostic and statistical manual of mental disorders, 4th ed. criteria.^[21,22] A study by Kim et al. in 2005, which used a single question (“Have you ever felt the need to move your legs or an unpleasant feeling like scary sensations in your legs before sleeping?”) reported a high prevalence using,^[23] but this method may overestimate RLS as it does not rule out other conditions.

Overall, RLS has a low prevalence in African countries, India, and several Asian countries, with higher rates observed in Northern Europe.^[24] The north-south gradient in prevalence may be due to genetic and methodological factors as well as environmental influences such as climate and ecology.

Several studies, including ours, have found a higher RLS prevalence in females and individuals over 50 years old. Tison et al. reported RLS in 10.8% of women and 5.8% of men, with prevalence increasing until 64 years before declining in both sexes.^[8] Allen et al. and Berger et al. also observed that RLS is twice as frequent in women, with a peak in prevalence up to age 79 before decreasing.^[25,26] This female predominance is likely linked to estrogen, with an increase in RLS during the third trimester of pregnancy, coinciding with an estrogen peak. Iron deficiency, which plays a well-known role in RLS pathogenesis and is common during pregnancy and menstruation, may also contribute.^[27,28] In addition, literature reviews show that, in addition to elderly subjects, RLS can affect children and adolescents though many studies, including ours, have not included these age groups.^[28]

RLS may also complicate diabetes mellitus,^[29,30] affecting 18% of diabetics, compared to 5.5% of non-diabetics.^[29] The risk of RLS was 1.79 times higher in diabetics,^[30] with a significative association between the syndrome and the diabète. Cho et al. found that diabetic individuals have a fourfold higher risk, which may be partly due to diabetic neuropathy.^[31] Some researchers suggest a link between RLS and cardiovascular disease, citing sleep disruption, sympathetic hyperactivity in RLS, and underlying periodic limb movements, all of which can contribute to HBP. The majority of RLS patients report 200–300 periodic limb movements per night.^[32] However, Elwood and Winter found this cardiovascular risk low.^[33,34] Gao et al. observed that the RLS prevalence increases with BMI and waist circumference,^[35] with an adjusted OR of 2.9 in our study. A separate cross-sectional study on 1803 individuals aged 18 and older found that each 5 kg/m² increase in BMI raised the likelihood of RLS by 31%.^[15] Obesity may contribute to RLS by reducing dopaminergic receptor sensitivity and the development of type 2 diabetes.^[36] In addition, both obesity and RLS share metabolic risk factors, and obese patients often suffer from vascular damage, further increasing RLS risk.^[34] HIV was not associated with RLS in this study. The relationship between HIV-associated neurodegeneration in central nervous system (CNS) sensorimotor modulation areas and RLS symptoms remains unclear, though longer HIV duration may lead to chronic CNS inflammation, iron deficiency in the CNS, and loss of myelin integrity.^[37]

A primary limitation of our study is that age was self-reported by illiterate participants, which could. introduce bias. As one of the first studies of its kind in Benin and West Africa, this study employed rigorous methodology based on validated criteria and random sampling.