Published Ahead of Print on February 26, 2016 as 10.1212/WNL.0000000000002521
Reduced bone resorption and increased bone mineral density in women with restless legs syndrome Mehmet Ali Cikrikcioglu, MD Yahya Sekin, MD Gulistan Halac, MD Elif Kilic, MD Siddika Kesgin, MD Senay Aydin, MD Nihal Ozaras, MD Onur Akan, MD Kenan Celik, MD Jamshid Hamdard, MD Mehmet Zorlu, MD Cumali Karatoprak, MD Mustafa Cakirca, MD Muharrem Kiskac, MD
Correspondence to Dr. Cikrikcioglu: [email protected]
ABSTRACT
Objective: To investigate bone resorption and formation markers as well as bone mineral density in women with restless legs syndrome (RLS).
Methods: This was a prospective cross-sectional case-control study involving drug-naive women with RLS and age- and body mass index (BMI)–matched female controls. Routine blood analyses, markers of bone formation, procollagen 1 n-terminal peptide, bone resorption, c-telopeptide of type 1 collagen (CTX), sclerostin, and bone mineral density (BMD) were compared between the 2 groups. Pregnant or breastfeeding women and individuals with comorbidities other than iron deficiency, type 2 diabetes mellitus, or hypertension were excluded.
Results: A significant increase in lumbar BMD was found among 78 women with RLS as compared to 78 age- and BMI-matched controls (p 5 0.001). The proportion of patients with osteopenia as defined by a lumbar T score was significantly lower among patients with RLS (p 5 0.040). CTX and sclerostin were significantly lower in patients with RLS (p 5 0.006 and p 5 0.011, respectively), as were the levels of 25-hydroxy vitamin D3, calcemia, and free T3 (p 5 0.017, p 5 0.017, and p 5 0.002, respectively).
Conclusions: Despite lower 25-hydroxy vitamin D3, patients with RLS had lower bone resorption markers, higher lumbar BMD, and lower frequency of lumbar osteopenia. As patients with RLS make movements night and day to decrease the severity of their symptoms, they unconsciously perform exercise, which may potentially explain the better bone profile among patients with RLS than in controls. Neurology® 2016;86:1–7 GLOSSARY BMD 5 bone mineral density; BMI 5 body mass index; CBC 5 complete blood count; CTX 5 C-telopeptide of type 1 collagen; IRLSSG 5 International Restless Legs Syndrome Study Group; PINP 5 procollagen 1 N-terminal peptide; RLS 5 restless legs syndrome.
Restless legs syndrome (RLS) is a sensory-motor disorder characterized by intense, unpleasant, and disagreeable sensations (paresthesia and dysesthesia) in the extremities (mostly in the legs) that tend to begin or worsen in the evening or at night during periods of rest and are relieved by movement.1 It affects nearly 5%–15% of adults.1 An increase in sympathetic nervous system activity as well as increased nocturnal urinary cortisol excretion have been reported in patients with RLS when compared with controls.2,3 Both cortisol and sympathetic nervous system are known to negatively affect bone metabolism.4,5 To our knowledge, no previous studies have examined bone resorption and formation markers and bone mineral density in patients with RLS. Thus, we analyzed telopeptide of type 1 collagen (CTX) and sclerostin (markers of bone resorption), procollagen 1 N-terminal peptide (PINP) (a marker of bone formation), and bonemineral density (BMD) in patients with RLS.6 METHODS Study design and study population. The study was designed as a prospective, cross-sectional, case-control study. Female patients with drug-naive idiopathic RLS consecutively admitted to The Neurology Clinic of Bezmialem Vakif University between April 2014 and March 2015 were enrolled in this study if they met the inclusion/exclusion criteria and provided written Editorial, page XXX From the Departments of Internal Medicine (M.A.C., Y.S., K.C., J.H., M.Z., C.K., M.C., M.K.), Neurology (G.H.), Biochemistry (E.K., S.K.), and Physical Medicine and Rehabilitation (N.O.), Bezmialem Vakif University Medical Faculty, Fatih; Department of Neurology (S.A.), Yedikule Chest Diseases and Chest Surgery Training and Research Hospital, Zeytinburnu; and Department of Neurology (O.A.), Okmeydani Training and Research Hospital, Sisli, Istanbul, Turkey. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article. © 2016 American Academy of Neurology
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
1
informed consent for this study. Similarly, age and body mass index (BMI)–matched women successively attending the internal medicine outpatient department during the same time period served as controls if they met the inclusion and exclusion criteria in addition to providing written informed consent.
Inclusion criteria. Drug-naive (i.e., no previous use of iron, dopamine agonists, opioids, or antiepileptics) female patients with RLS between 18 and 75 years of age meeting the 4 diagnostic criteria proposed by the International Restless Legs Syndrome Study Group (IRLSSG) and diagnosed with RLS with or without iron deficiency anemia comprised the patient group.7 The control group consisted of RLS-free women, with or without iron deficiency anemia.
Exclusion criteria. Exclusion criteria included the following conditions: pregnancy, lactation, chronic kidney disease $ stage 3, diagnosis of mood disorder or psychosis, use of antidepressant or antipsychotic drugs or lithium, antiepileptic or opioid or dopaminergic drug use, alcoholism, drug addiction, hypophyseal dysfunction, thyroid dysfunction, primary parathyroid dysfunction, adrenal dysfunction, use of oral contraceptives, postmenopausal hormone replacement therapy, being bedridden, presence of bone prosthesis or orthosis, major surgical intervention, history of bone fracture, chronic and acute infections, cerebrovascular accident, ischemic heart disease, heart failure, chronic obstructive lung disease, asthma, obstructive sleep apnea syndrome, malabsorption syndromes, dementia, Parkinson disease, multiple sclerosis, inflammatory bowel disease, collagen tissue disorders, sarcoidosis, chronic hepatitis, portal hypertension, chronic corticosteroid and immunosuppressive therapy, hemolytic anemias, hematologic malignancy, and solid tumors. Blood sampling and analysis. Venous blood sampling was carried out in the early morning hours after overnight fasting. On the same day, complete blood count (CBC), biochemistry, 25-hydroxy vitamin D3, parathyroid hormone, ferritin, and thyroid hormone analyses were performed. For ELISA assay, venous blood samples were left for 20 minutes to allow clotting and then centrifuged at 1,500 g for 15 minutes. The sera were stored at 280°C for 12 months and thawed at room temperature at the day of analysis. CBC analysis was done with Sysmex XT 1800i apparatus (Roche, Kobe, Japan). Biochemical analyses were carried out using a COBAS 8000 apparatus (Roche, Tokyo, Japan) and COBAS-C system kits. The 25-hydroxy vitamin D3 analysis was carried out with COBAS E 601 hormone analyzer (Roche) using Roche kits (Mannheim, Germany). Thyroid hormone levels were analyzed with Advia (Tarrytown, NY) Centaur kit, and parathormone levels were analyzed with intact parathormone kit (Bayswater, Victoria, Australia). Ferritin levels were analyzed with ferritin kits (Bayswater) using an Advia (Dublin, Ireland) Centaur device. The serum concentrations of human PINP were measured with ELISA kit, according to protocols provided by the manufacturers (Human PINP ELISA kit, ref no.: DZE201121351; lot no.: 201410; Sunred Biological Technology, Shanghai, China). Serum concentrations of human CTX were measured with ELISA kit, according to protocols provided by the manufacturers (Human CTX-1 ELISA kit, ref no.: DZE201121350; lot no.: 201410; Sunred Biological Technology). Serum concentrations of human sclerostin were measured with ELISA kit, in line with the manufacturer’s recommended protocols (Human SOST Elisa Kit, ref no.: DZE201125418; lot no.: 201503; Sunred Biological Technology). Multiskan FC Microplate Photometer (Thermo Fisher Scientific, Waltham, MA) was used for reading at 450 nm. The ELISA results were expressed in ng/mL. 2
Neurology 86
Measurement of BMD. BMD was measured at L1-L4 lumbar vertebra and in left proximal femur using dual energy x-ray absorptiometry technique with a DPX–Lunar trademark device (General Electric Healthcare, Cleveland, OH). Patient assessments. Medical history and family history were obtained from each person enrolled in the study. The time before the patient had been diagnosed with RLS or the period in which each patient had RLS without any treatment was questioned (duration of disease). The severity of RLS symptoms was determined via the John Hopkins RLS Severity Scale and the RLS Severity Scale, which was developed by the IRLSSG.8,9 Smoking status, age at menarche, menopause status (spontaneous or surgical menopause, age at menopause), number of miscarriages or curettages, duration of breastfeeding, number of births, monthly income of family, education status, employment status (employed vs not employed), and status of retirement were determined. A systemic physical examination was performed in each study participant. Regular use of medications and comorbidities were recorded. Also, arterial blood pressure was measured in all subjects from the right arm using a mechanical sphygmomanometer after 15 minutes of rest in the early morning hours, i.e., before blood samples were obtained. BMI was calculated using the following formula: BMI 5 body weight (kg)/height 3 height (m). Statistical methods. Numerical variables were presented as means with standard deviation, and nominal variables were presented as ratios. Participants of the study were divided into RLS and control groups. Nominal independent variables were compared between the groups using the x2 test. One-sample Kolmogorov-Smirnov test was performed to determine if the continuous (numerical) independent variables were normally distributed or not. In comparison of groups, independent continuous variables with normal distribution were compared with Student t test, whereas independent continuous variables without normal distribution were compared with MannWhitney U test. Bivariate correlations were sought and linear regression analysis was also performed. Power calculations for the study were performed after completion of study procedures. A 2-tailed p value of ,0.05 was considered to be statistically significant. Standard protocol approvals, registrations, and patient consents. The study was approved by Bezmialem Vakif University Medical Faculty Ethics Committee, and all participants recruited to the study gave informed consent. The study was performed in accordance with the 2009 Helsinki Declaration. RESULTS There were 78 female patients with RLS and 78 female controls in the study. The mean age was 45.7 6 11.7 years (range 21–75). The mean BMI was 28.6 6 4.5 kg/m2 (range 19.6–40). The 2 groups were similar in terms of age and BMI (table 1). The age at menarche, proportion of individuals at menopause, age at menopause, frequency of surgical menopause, number of births, number of miscarriages or curettages, and total duration of breastfeeding were comparable across the 2 groups (tables 1 and 2). The 2 groups were similar in terms of monthly family income, education status, employment status (employed vs not employed), and retirement status (tables 1 and 2).
March 29, 2016
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
Table 1
Comparison of groups in terms of demographic variables, blood pressure, RLS severity score, and disease duration
Variables
RLS (78 subjects)
Controls (78 subjects)
p Value
Type of distribution
Age, y
46.45 6 11.26
45.01 6 12.27
0.441
ND
BMI, kg/m2
28.80 6 4.61
28.48 6 4.37
0.654
ND
Births, n
2.54 6 1.34
2.63 6 1.55
0.785
NND
Miscarriage or curettage, n
0.70 6 1.11
0.63 6 1.08
0.706
NND
Age at menarche, y
13.45 6 1.63
13.19 6 1.17
0.294
NND
Age at menopause, y
46.48 6 4.82
48.39 6 5.26
0.179
ND
Duration of breastfeeding, mo
32.33 6 27.01
33.74 6 28.60
0.834
NND
Monthly income, Turkish Lira
2,348.8 6 1,535.6
2,156.6 6 1,105.7
0.676
NND
Systolic blood pressure, mm Hg
123.86 6 12.74
125.60 6 12.56
0.401
ND
Diastolic blood pressure, mm Hg
70.11 6 7.72
70.96 6 8.90
IRLSSG RLS severity score, n
28.9 6 5.2
0 6 00
,0.001
NND
Duration of disease, mo
45 6 44.5
0 6 00
,0.001
NND
0.535
ND
Abbreviations: BMI 5 body mass index; IRLSSG 5 International Restless Legs Syndrome Study Group; ND 5 normal distribution; NND 5 non-normal distribution; RLS 5 restless legs syndrome.
There were no significant differences between groups in terms of the proportion of cigarette smokers and ex-smokers, blood pressures, occurrence of hypertension or diabetes, HbA1C and fasting blood glucose levels, and regular use of medications (tables 1–3). Patients with RLS had significantly lower blood calcium, 25-hydroxy vitamin D3, sclerostin, and CTX and higher lumbar BMD than controls (table 3). PINP and femur BMD were comparable between the 2 groups (table 3). Significant deficiency of 25-hydroxy vitamin D3 was detected in 83% of the overall study population (table 2). Although more patients in the RLS group had 25-hydroxy vitamin D3 deficiency, the difference was not significant (table 2). There were significantly fewer individuals with osteopenia, as reflected by the lumbar T scores, among patients with RLS (table 2). A similar number of individuals in both groups had osteoporosis according to lumbar and femoral T scores, and a similar number of individuals in both groups had osteopenia according to the femoral T score (table 2). Thyroid-stimulating hormone and FT4 were similar in the study groups, although FT3 was significantly lower in the RLS group (table 3). No significant differences could be detected between the 2 groups with respect to the results of other routine blood analyses (table 3). According to Johns Hopkins RLS severity criteria, mild or moderate severity was present in 75% of the patients, while the patient group consisted of participants with moderately severe RLS with regard to IRLSSG RLS severity score (tables 1 and 2). The mean duration of disease was 45 6 44.5 months (range 8–360 months).
There was a negative bivariate correlation between IRLSSG severity score and calcemia, between IRLSSG severity score and CTX, and between IRLSSG severity score and sclerostin (r 5 20.173, p 5 0.029; r 5 20.244, p 5 0.002; and r 5 20.271, p 5 0.010, respectively). There was a positive bivariate correlation between IRLSSG severity score and lumbar BMD (r 5 0.214, p 5 0.007). A negative bivariate correlation between the duration of disease and CTX, between the duration of disease and sclerostin, and between the duration of disease and calcemia (r 5 20.185, p 5 0.020; r 5 20.171, p 5 0.050; and r 5 20.190, p 5 0.016, respectively) was observed. There was positive bivariate correlation between the duration of disease and lumbar BMD (r 5 0.190, p 5 0.018). In linear regression analysis, while there was no correlation between the dependent variable, i.e., lumbar BMD, and the independent variables, i.e., age at menarche, number of births, number of miscarriages or curettage, duration of breastfeeding, employment status (employed vs not employed), cigarette smoking, calcemia, phosphatemia, creatinine, albumin, 25-hydroxy vitamin D3, transferrin saturation rate, duration of disease, and IRLSSG severity score, a significant correlation between the dependent variable, i.e., lumbar BMD, and the independent variables, i.e., presence of RLS, presence of menopause, BMI, and monthly family income, was found (table 4). In linear regression analysis, no correlation between the dependent variable, i.e., calcemia, and the independent variables, i.e., creatinine, 25hydroxy vitamin D3 level, C-reactive protein, HbA1c, duration of disease, IRLSSG RLS severity score, number of births, number of miscarriages or curettage, Neurology 86
March 29, 2016
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
3
Table 2
Demographic features, medicine regularly used, RLS severity, vitamin D, and t score status among the participants of the study
Variable
RLS (78 subjects), n
Controls (78 subjects), n p Value
Menopause
29
23
0.396
Surgical menopause
3
4
1
Diabetes mellitus type 2
6
9
0.587
Essential hypertension
12
15
0.672
Smokers
27
27
0.866
Ex-smokers
5
3
0.717
Illiterate
9
12
0.639
Primary school graduate
45
44
1
Secondary school graduate
8
6
0.779
High school graduate
8
7
1
University graduate
8
9
1
Unemployed
60
65
0.422
Working
11
10
1
Student
2
0
0.477
Retired
5
3
0.717
Using oral antidiabetic medicine
5
8
0.562
Using insulin and oral antidiabetic medicine
1
1
0.477
Using antihypertensive medicine
11
13
0.868
Women with mild RLS according to Johns Hopkins severity scale
24
0
,0.001
Women with moderate RLS according to Johns Hopkins severity scale
36
0
,0.001
Women with severe RLS according to Johns Hopkins severity scale
18
0
,0.001
Women with lumbar t score £22.5 (osteoporosis)
3
3
0.677
Women with 22.5
Reduced bone resorption and increased bone mineral density in women with restless legs syndrome Mehmet Ali Cikrikcioglu, MD Yahya Sekin, MD Gulistan Halac, MD Elif Kilic, MD Siddika Kesgin, MD Senay Aydin, MD Nihal Ozaras, MD Onur Akan, MD Kenan Celik, MD Jamshid Hamdard, MD Mehmet Zorlu, MD Cumali Karatoprak, MD Mustafa Cakirca, MD Muharrem Kiskac, MD
Correspondence to Dr. Cikrikcioglu: [email protected]
ABSTRACT
Objective: To investigate bone resorption and formation markers as well as bone mineral density in women with restless legs syndrome (RLS).
Methods: This was a prospective cross-sectional case-control study involving drug-naive women with RLS and age- and body mass index (BMI)–matched female controls. Routine blood analyses, markers of bone formation, procollagen 1 n-terminal peptide, bone resorption, c-telopeptide of type 1 collagen (CTX), sclerostin, and bone mineral density (BMD) were compared between the 2 groups. Pregnant or breastfeeding women and individuals with comorbidities other than iron deficiency, type 2 diabetes mellitus, or hypertension were excluded.
Results: A significant increase in lumbar BMD was found among 78 women with RLS as compared to 78 age- and BMI-matched controls (p 5 0.001). The proportion of patients with osteopenia as defined by a lumbar T score was significantly lower among patients with RLS (p 5 0.040). CTX and sclerostin were significantly lower in patients with RLS (p 5 0.006 and p 5 0.011, respectively), as were the levels of 25-hydroxy vitamin D3, calcemia, and free T3 (p 5 0.017, p 5 0.017, and p 5 0.002, respectively).
Conclusions: Despite lower 25-hydroxy vitamin D3, patients with RLS had lower bone resorption markers, higher lumbar BMD, and lower frequency of lumbar osteopenia. As patients with RLS make movements night and day to decrease the severity of their symptoms, they unconsciously perform exercise, which may potentially explain the better bone profile among patients with RLS than in controls. Neurology® 2016;86:1–7 GLOSSARY BMD 5 bone mineral density; BMI 5 body mass index; CBC 5 complete blood count; CTX 5 C-telopeptide of type 1 collagen; IRLSSG 5 International Restless Legs Syndrome Study Group; PINP 5 procollagen 1 N-terminal peptide; RLS 5 restless legs syndrome.
Restless legs syndrome (RLS) is a sensory-motor disorder characterized by intense, unpleasant, and disagreeable sensations (paresthesia and dysesthesia) in the extremities (mostly in the legs) that tend to begin or worsen in the evening or at night during periods of rest and are relieved by movement.1 It affects nearly 5%–15% of adults.1 An increase in sympathetic nervous system activity as well as increased nocturnal urinary cortisol excretion have been reported in patients with RLS when compared with controls.2,3 Both cortisol and sympathetic nervous system are known to negatively affect bone metabolism.4,5 To our knowledge, no previous studies have examined bone resorption and formation markers and bone mineral density in patients with RLS. Thus, we analyzed telopeptide of type 1 collagen (CTX) and sclerostin (markers of bone resorption), procollagen 1 N-terminal peptide (PINP) (a marker of bone formation), and bonemineral density (BMD) in patients with RLS.6 METHODS Study design and study population. The study was designed as a prospective, cross-sectional, case-control study. Female patients with drug-naive idiopathic RLS consecutively admitted to The Neurology Clinic of Bezmialem Vakif University between April 2014 and March 2015 were enrolled in this study if they met the inclusion/exclusion criteria and provided written Editorial, page XXX From the Departments of Internal Medicine (M.A.C., Y.S., K.C., J.H., M.Z., C.K., M.C., M.K.), Neurology (G.H.), Biochemistry (E.K., S.K.), and Physical Medicine and Rehabilitation (N.O.), Bezmialem Vakif University Medical Faculty, Fatih; Department of Neurology (S.A.), Yedikule Chest Diseases and Chest Surgery Training and Research Hospital, Zeytinburnu; and Department of Neurology (O.A.), Okmeydani Training and Research Hospital, Sisli, Istanbul, Turkey. Go to Neurology.org for full disclosures. Funding information and disclosures deemed relevant by the authors, if any, are provided at the end of the article. © 2016 American Academy of Neurology
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
1
informed consent for this study. Similarly, age and body mass index (BMI)–matched women successively attending the internal medicine outpatient department during the same time period served as controls if they met the inclusion and exclusion criteria in addition to providing written informed consent.
Inclusion criteria. Drug-naive (i.e., no previous use of iron, dopamine agonists, opioids, or antiepileptics) female patients with RLS between 18 and 75 years of age meeting the 4 diagnostic criteria proposed by the International Restless Legs Syndrome Study Group (IRLSSG) and diagnosed with RLS with or without iron deficiency anemia comprised the patient group.7 The control group consisted of RLS-free women, with or without iron deficiency anemia.
Exclusion criteria. Exclusion criteria included the following conditions: pregnancy, lactation, chronic kidney disease $ stage 3, diagnosis of mood disorder or psychosis, use of antidepressant or antipsychotic drugs or lithium, antiepileptic or opioid or dopaminergic drug use, alcoholism, drug addiction, hypophyseal dysfunction, thyroid dysfunction, primary parathyroid dysfunction, adrenal dysfunction, use of oral contraceptives, postmenopausal hormone replacement therapy, being bedridden, presence of bone prosthesis or orthosis, major surgical intervention, history of bone fracture, chronic and acute infections, cerebrovascular accident, ischemic heart disease, heart failure, chronic obstructive lung disease, asthma, obstructive sleep apnea syndrome, malabsorption syndromes, dementia, Parkinson disease, multiple sclerosis, inflammatory bowel disease, collagen tissue disorders, sarcoidosis, chronic hepatitis, portal hypertension, chronic corticosteroid and immunosuppressive therapy, hemolytic anemias, hematologic malignancy, and solid tumors. Blood sampling and analysis. Venous blood sampling was carried out in the early morning hours after overnight fasting. On the same day, complete blood count (CBC), biochemistry, 25-hydroxy vitamin D3, parathyroid hormone, ferritin, and thyroid hormone analyses were performed. For ELISA assay, venous blood samples were left for 20 minutes to allow clotting and then centrifuged at 1,500 g for 15 minutes. The sera were stored at 280°C for 12 months and thawed at room temperature at the day of analysis. CBC analysis was done with Sysmex XT 1800i apparatus (Roche, Kobe, Japan). Biochemical analyses were carried out using a COBAS 8000 apparatus (Roche, Tokyo, Japan) and COBAS-C system kits. The 25-hydroxy vitamin D3 analysis was carried out with COBAS E 601 hormone analyzer (Roche) using Roche kits (Mannheim, Germany). Thyroid hormone levels were analyzed with Advia (Tarrytown, NY) Centaur kit, and parathormone levels were analyzed with intact parathormone kit (Bayswater, Victoria, Australia). Ferritin levels were analyzed with ferritin kits (Bayswater) using an Advia (Dublin, Ireland) Centaur device. The serum concentrations of human PINP were measured with ELISA kit, according to protocols provided by the manufacturers (Human PINP ELISA kit, ref no.: DZE201121351; lot no.: 201410; Sunred Biological Technology, Shanghai, China). Serum concentrations of human CTX were measured with ELISA kit, according to protocols provided by the manufacturers (Human CTX-1 ELISA kit, ref no.: DZE201121350; lot no.: 201410; Sunred Biological Technology). Serum concentrations of human sclerostin were measured with ELISA kit, in line with the manufacturer’s recommended protocols (Human SOST Elisa Kit, ref no.: DZE201125418; lot no.: 201503; Sunred Biological Technology). Multiskan FC Microplate Photometer (Thermo Fisher Scientific, Waltham, MA) was used for reading at 450 nm. The ELISA results were expressed in ng/mL. 2
Neurology 86
Measurement of BMD. BMD was measured at L1-L4 lumbar vertebra and in left proximal femur using dual energy x-ray absorptiometry technique with a DPX–Lunar trademark device (General Electric Healthcare, Cleveland, OH). Patient assessments. Medical history and family history were obtained from each person enrolled in the study. The time before the patient had been diagnosed with RLS or the period in which each patient had RLS without any treatment was questioned (duration of disease). The severity of RLS symptoms was determined via the John Hopkins RLS Severity Scale and the RLS Severity Scale, which was developed by the IRLSSG.8,9 Smoking status, age at menarche, menopause status (spontaneous or surgical menopause, age at menopause), number of miscarriages or curettages, duration of breastfeeding, number of births, monthly income of family, education status, employment status (employed vs not employed), and status of retirement were determined. A systemic physical examination was performed in each study participant. Regular use of medications and comorbidities were recorded. Also, arterial blood pressure was measured in all subjects from the right arm using a mechanical sphygmomanometer after 15 minutes of rest in the early morning hours, i.e., before blood samples were obtained. BMI was calculated using the following formula: BMI 5 body weight (kg)/height 3 height (m). Statistical methods. Numerical variables were presented as means with standard deviation, and nominal variables were presented as ratios. Participants of the study were divided into RLS and control groups. Nominal independent variables were compared between the groups using the x2 test. One-sample Kolmogorov-Smirnov test was performed to determine if the continuous (numerical) independent variables were normally distributed or not. In comparison of groups, independent continuous variables with normal distribution were compared with Student t test, whereas independent continuous variables without normal distribution were compared with MannWhitney U test. Bivariate correlations were sought and linear regression analysis was also performed. Power calculations for the study were performed after completion of study procedures. A 2-tailed p value of ,0.05 was considered to be statistically significant. Standard protocol approvals, registrations, and patient consents. The study was approved by Bezmialem Vakif University Medical Faculty Ethics Committee, and all participants recruited to the study gave informed consent. The study was performed in accordance with the 2009 Helsinki Declaration. RESULTS There were 78 female patients with RLS and 78 female controls in the study. The mean age was 45.7 6 11.7 years (range 21–75). The mean BMI was 28.6 6 4.5 kg/m2 (range 19.6–40). The 2 groups were similar in terms of age and BMI (table 1). The age at menarche, proportion of individuals at menopause, age at menopause, frequency of surgical menopause, number of births, number of miscarriages or curettages, and total duration of breastfeeding were comparable across the 2 groups (tables 1 and 2). The 2 groups were similar in terms of monthly family income, education status, employment status (employed vs not employed), and retirement status (tables 1 and 2).
March 29, 2016
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
Table 1
Comparison of groups in terms of demographic variables, blood pressure, RLS severity score, and disease duration
Variables
RLS (78 subjects)
Controls (78 subjects)
p Value
Type of distribution
Age, y
46.45 6 11.26
45.01 6 12.27
0.441
ND
BMI, kg/m2
28.80 6 4.61
28.48 6 4.37
0.654
ND
Births, n
2.54 6 1.34
2.63 6 1.55
0.785
NND
Miscarriage or curettage, n
0.70 6 1.11
0.63 6 1.08
0.706
NND
Age at menarche, y
13.45 6 1.63
13.19 6 1.17
0.294
NND
Age at menopause, y
46.48 6 4.82
48.39 6 5.26
0.179
ND
Duration of breastfeeding, mo
32.33 6 27.01
33.74 6 28.60
0.834
NND
Monthly income, Turkish Lira
2,348.8 6 1,535.6
2,156.6 6 1,105.7
0.676
NND
Systolic blood pressure, mm Hg
123.86 6 12.74
125.60 6 12.56
0.401
ND
Diastolic blood pressure, mm Hg
70.11 6 7.72
70.96 6 8.90
IRLSSG RLS severity score, n
28.9 6 5.2
0 6 00
,0.001
NND
Duration of disease, mo
45 6 44.5
0 6 00
,0.001
NND
0.535
ND
Abbreviations: BMI 5 body mass index; IRLSSG 5 International Restless Legs Syndrome Study Group; ND 5 normal distribution; NND 5 non-normal distribution; RLS 5 restless legs syndrome.
There were no significant differences between groups in terms of the proportion of cigarette smokers and ex-smokers, blood pressures, occurrence of hypertension or diabetes, HbA1C and fasting blood glucose levels, and regular use of medications (tables 1–3). Patients with RLS had significantly lower blood calcium, 25-hydroxy vitamin D3, sclerostin, and CTX and higher lumbar BMD than controls (table 3). PINP and femur BMD were comparable between the 2 groups (table 3). Significant deficiency of 25-hydroxy vitamin D3 was detected in 83% of the overall study population (table 2). Although more patients in the RLS group had 25-hydroxy vitamin D3 deficiency, the difference was not significant (table 2). There were significantly fewer individuals with osteopenia, as reflected by the lumbar T scores, among patients with RLS (table 2). A similar number of individuals in both groups had osteoporosis according to lumbar and femoral T scores, and a similar number of individuals in both groups had osteopenia according to the femoral T score (table 2). Thyroid-stimulating hormone and FT4 were similar in the study groups, although FT3 was significantly lower in the RLS group (table 3). No significant differences could be detected between the 2 groups with respect to the results of other routine blood analyses (table 3). According to Johns Hopkins RLS severity criteria, mild or moderate severity was present in 75% of the patients, while the patient group consisted of participants with moderately severe RLS with regard to IRLSSG RLS severity score (tables 1 and 2). The mean duration of disease was 45 6 44.5 months (range 8–360 months).
There was a negative bivariate correlation between IRLSSG severity score and calcemia, between IRLSSG severity score and CTX, and between IRLSSG severity score and sclerostin (r 5 20.173, p 5 0.029; r 5 20.244, p 5 0.002; and r 5 20.271, p 5 0.010, respectively). There was a positive bivariate correlation between IRLSSG severity score and lumbar BMD (r 5 0.214, p 5 0.007). A negative bivariate correlation between the duration of disease and CTX, between the duration of disease and sclerostin, and between the duration of disease and calcemia (r 5 20.185, p 5 0.020; r 5 20.171, p 5 0.050; and r 5 20.190, p 5 0.016, respectively) was observed. There was positive bivariate correlation between the duration of disease and lumbar BMD (r 5 0.190, p 5 0.018). In linear regression analysis, while there was no correlation between the dependent variable, i.e., lumbar BMD, and the independent variables, i.e., age at menarche, number of births, number of miscarriages or curettage, duration of breastfeeding, employment status (employed vs not employed), cigarette smoking, calcemia, phosphatemia, creatinine, albumin, 25-hydroxy vitamin D3, transferrin saturation rate, duration of disease, and IRLSSG severity score, a significant correlation between the dependent variable, i.e., lumbar BMD, and the independent variables, i.e., presence of RLS, presence of menopause, BMI, and monthly family income, was found (table 4). In linear regression analysis, no correlation between the dependent variable, i.e., calcemia, and the independent variables, i.e., creatinine, 25hydroxy vitamin D3 level, C-reactive protein, HbA1c, duration of disease, IRLSSG RLS severity score, number of births, number of miscarriages or curettage, Neurology 86
March 29, 2016
ª 2016 American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
3
Table 2
Demographic features, medicine regularly used, RLS severity, vitamin D, and t score status among the participants of the study
Variable
RLS (78 subjects), n
Controls (78 subjects), n p Value
Menopause
29
23
0.396
Surgical menopause
3
4
1
Diabetes mellitus type 2
6
9
0.587
Essential hypertension
12
15
0.672
Smokers
27
27
0.866
Ex-smokers
5
3
0.717
Illiterate
9
12
0.639
Primary school graduate
45
44
1
Secondary school graduate
8
6
0.779
High school graduate
8
7
1
University graduate
8
9
1
Unemployed
60
65
0.422
Working
11
10
1
Student
2
0
0.477
Retired
5
3
0.717
Using oral antidiabetic medicine
5
8
0.562
Using insulin and oral antidiabetic medicine
1
1
0.477
Using antihypertensive medicine
11
13
0.868
Women with mild RLS according to Johns Hopkins severity scale
24
0
,0.001
Women with moderate RLS according to Johns Hopkins severity scale
36
0
,0.001
Women with severe RLS according to Johns Hopkins severity scale
18
0
,0.001
Women with lumbar t score £22.5 (osteoporosis)
3
3
0.677
Women with 22.5
