Free AccessLight

# Decreased Symptoms without Augmented Skin Blood Flow in Subjects with RLS/WED after Vibration Treatment

Published Online:https://doi.org/10.5664/jcsm.5920Cited by:9

## ABSTRACT

### Study Objectives:

Vascular disturbances leading to tissue hypoxia have been named as a possible cause for RLS/WED. Vibration to the whole body (WBV) in subjects with RLS/WED results in increased skin blood flow (SBF). The aims of this investigation were to (1) determine if a two-week treatment with WBV will decrease symptoms associated with RLS/WED and, (2) if so, determine if the mechanism for improvement in symptoms is related to an increase in SBF, as measured in flux.

### Methods:

Eleven subjects with RLS/WED underwent 2 weeks of 14-minute intermittent WBV and a 2-week sham treatment in randomized order. Pre and post intervention RLS symptom severity were compared. Baseline SBF was compared between subjects with RLS/WED and an age- and sex-matched control group. A crossover design (aim 1) and a matched case-control design and repeated measures design (aim 2) were used. The data analyses consisted of 2-sample and paired t-tests; where applicable we used a standard crossover design adjustment.

### Results:

WBV did significantly decrease symptoms associated with RLS/WED compared to baseline data and compared to sham treatment. The baseline flux was significantly lower in RLS/WED subjects than matched controls, but this deficit was negated with WBV. There was no increase in resting SBF over the 2 weeks of treatment.

### Conclusions:

Subjects with RLS/WED have decreased SBF but are able to increase flux to the same level as normal subjects with WBV. A 2-week intervention with WBV decreases symptoms associated with RLS/WED, but this does not seem to be related to an increase in resting SBF.

### Citation:

Mitchell UH, Hilton SC, Hunsaker E, Ulfberg J. Decreased symptoms without augmented skin blood flow in subjects with RLS/WED after vibration treatment. J Clin Sleep Med 2016;12(7):947–952.

## INTRODUCTION

Restless legs syndrome (RLS), or now known by its newer name Willis-Ekbom Disease (WED) is a chronic sensorimotor disorder, characterized by a strong urge to move.1 This urge is usually felt in the legs, but can occur anywhere else in the body as well. It is triggered or accompanied by an uncomfortable feeling which is often described as a “creeping, tugging, pulling.”1 The symptoms often become worse as the day progresses, and the symptoms are usually lessened by movement.2 The pathogenesis of RLS/WED is not clear, but there exist several hypotheses. The most commonly mentioned and most researched one involves the neurotransmitter dopamine in the brain.3

The theory behind our current line of research is based on the historical pathophysiology model that postulates that decreased peripheral blood flow leads to hypoxia which is thought to be at least partially responsible for the characteristic unpleasant sensations and the urge to move.47 This oxygen-based vascular theory has gained more traction in recent years.811 Different circulation-enhancing modalities, such as pneumatic compression devices46 and near-infrared light7,12 have been used for the treatment of RLS/WED and found to be successful in decreasing symptoms and improving quality of life. The tissue hypoxia theory is supported by two other facts: (1) physical activity or movement lessens the symptoms associated with RLS/WED, and (2) it has been shown that lack of exercise is a strong predictor of and a significant risk factor for RLS/WED.11,13

### BRIEF SUMMARY

Current Knowledge/Study Rationale: Tissue hypoxia has been named as a possible cause for RLS/WED, and whole body vibration is known to increase blood flow and thus decrease tissue hypoxia. This study examined the possibility of using whole body vibration as a treatment option for RLS/WED.

Study Impact: This study offers a new short term treatment option for symptoms associated with RLS/WED. It also opens new doors into further research investigating the mechanism behind the success of this modality in reducing RLS/WED symptoms.

Externally applied low frequency oscillation or vibration to the whole body has been shown to lead to increased skin blood flow (SBF) in healthy individuals14 and to an even greater augmentation of superficial blood flow in subjects with RLS/ WED.11 In the aforementioned study it was found that baseline blood flow in RLS/WED subjects was significantly higher than that of controls' and that vibration yielded a significantly greater increase in blood flow in RLS/WED subjects when compared to the controls. Thus, the results indicated that subjects with RLS/WED have some kind of issue involving peripheral blood flow. Those data were collected in the early afternoon, still leaving the possibility that peripheral SBF could change in the evening, when the RLS/WED symptoms become more apparent. SBF as a potential marker for peripheral sympathetic nervous system activity and as an outcome measure for vasomotor reactivity has been used before.15

Research has yet to be conducted to determine if individuals with RLS/WED respond to treatment with whole body vibration (WBV) with decreases in RLS/WED symptoms and a more “normal” flux when compared to controls. The results could potentially serve as another piece in the pathogenesis puzzle of RLS/WED and lend themselves to create an alternative treatment option for sufferers of RLS/WED. WBV devices can be found in many fitness centers, as they can be used as part of a workout program, or they can be purchased for home use; they cost between $200 and$2,500.16

The aims of this investigation were to answer the following questions: (1) does a 2-week treatment with WBV decrease symptoms associated with RLS/WED as measured by the International RLS Study Group Rating Scale (IRLS)17 and RLS-6 scales18?, and (2) if so, is the mechanism for improvement in symptoms due to an increase in superficial peripheral SBF? This study was approved by the institutional review board.

## METHODS

A crossover design was used to collect data to answer research question (1) A matched case-control design and repeated measures design were used to collect data from the RLS/WED subjects to answer research question. (2) The data analyses consisted of 2-sample and paired t-tests; where applicable we used a standard cross-over design adjustment.19

For our power analysis we used a standard deviation of 5 points for the frequently used IRLS.17 This is a questionnaire that measures RLS severity on a 0–40 scale, using the answers to 10 questions, each of them giving 5 answer options: from 0 to 4. The more intense the symptoms, the higher the number attached to the answer. With a standard deviation of 5 points, an alpha of 0.05, and a beta of 0.2 (power of 0.8) we calculated that we needed 12 subjects to detect a difference in IRLS scores of 6 points.

Thirteen subjects with diagnosed RLS/WED were enrolled in this study. Inclusion criteria included a confirmed diagnosis of RLS/WED using the essential diagnostic criteria,17 for which all 5 questions pertaining to RLS/WED symptoms have to be answered affirmatively, and a severity score ≥ 11 on the IRLS. Exclusion criteria included presence of any one of the following: diabetic neuropathy, deep vein thrombosis, low blood iron, and open wound in the lower extremity or fracture anywhere in the body within the last 3 months. In order to answer the second research question, 11 control subjects were also enrolled in this study. Inclusion criteria for the control subjects included age- and sex-matching to the RLS/WED subjects, and a confirmed non-diagnosis of RLS/WED using the essential diagnostic criteria.”17 The exclusion criteria for the controls were the same as for the RLS/WED subjects.

Absence of neuropathy was confirmed by Quantitative Sensory Testing (QST) for heat and cold sensation thresholds, using the methods of limits procedure.20 We used the results of this test to exclude subjects with signs of neuropathy in small A-delta and C nerve fibers.20

Absence of deep vein thrombosis was confirmed by performing a compression ultrasound (portable ultrasound system from GE Healthcare, Hertfordshire, UK) bilaterally on the common femoral and the popliteal veins. This was done in order to rule out the threat of dislodging a clot during vibration. Information on blood iron content was obtained by interviewing the subjects; the presence of an open wound and fracture was assessed by inspection.

SBF of the dorsum of the right foot was measured using Laser Doppler imaging (Moor Instruments, Inc., Oxford, England). The image was captured and at a later point in time evaluated in processing software (Moor LDI Processing V3.1, Moor Instruments Inc., Oxford, England).21 The laser Doppler image (LDI) determines the velocity of red blood cells in arterioles, venules, and capillaries 1–4 mm under the skin surface. The output is the “flux” of red blood cells, defined as the number of red blood cells times their velocity, (measured in blood perfusion units [BPU]), which determines circulation.22 With the subject seated, a Doppler image was taken. The subjects transferred onto the vibration platform, located also in the chamber and underwent the vibration treatment or sham treatment in the order as dictated by group assignment and as described below. Immediately following the treatment, the subject transferred back to the chair and another LDI scan was performed. Each subject received SBF measurements before (“pre,” or “resting”) and after (“post”) vibration and sham treatments.

All RLS/WED subjects received WBV and sham treatment for 2 weeks, 3 times a week each. This frequency was chosen to mimic other studies that investigated circulation-enhancing modalities, such as exercise23 and near-infrared light.12,24 An actual treatment time of 2 weeks, and thus a subject commitment of 4 weeks, was chosen to increase our subject retention rate. It is known that the number of subjects dropping out of studies increases over time.25 All subjects received their treatments in the evening, between 18:00 and 22:00.

### Vibration Treatment

The subject stood on the vibration platform with knees flexed to about 40 degrees, the weight centered over the middle of each foot. Ten 30-s bouts of WBV at a frequency and amplitude of 26 Hz and 2 mm, respectively, was performed with a one-minute rest while standing between each bout.11

### Sham Treatment

The subject stood on the vibration platform assuming the same position, for the same intervals as above, but no vibration stimulus was given.11 This intervention was utilized to address the possibility that positioning alone can influence blood flow.

### Study Flow

RLS/WED subjects: On the first visit subjects completed two pre-participation questionnaires (demographics, medicine in-take, and confirmation of the RLS/WED diagnosis by asking the 5 essential diagnostic criteria17), and filled out the IRLS17 and RLS-6 scales.18 The IRLS is a 10-item questionnaire that was developed and validated in 2003.26 The RLS-6 scales use digital visual analog scales (VAS) to answer 6 questions (see supplemental material). A total score was not calculated across these 6 items; rather, each item produced its own score using 0–10.

Eligible subjects consequently underwent neuropathy20 and deep vein thrombosis screening. Upon meeting all inclusion criteria and after signing the informed consent form, the subjects drew a piece of paper out of an opaque envelope with their group assignments. They were: group A “vibration first” and group B “sham first.” The subjects then scheduled their appointments for the next 4 weeks.

For each treatment day, the subjects came to the Human Performance laboratory. After every last treatment day in the week the subjects filled out the IRLS17 and RLS-6 scales.18

Subjects with RLS/WED were recruited first. Once all subjects were enrolled in the study, we recruited age and sex-matched control subjects. The control subjects also underwent testing to rule out neuropathy and deep vein thrombosis and to confirm that they did not have RLS/WED. They came to the laboratory only once for a one-time vibration treatment session in order to measure the impact of WBV on their SBF or flux.

## RESULTS

### Subjects

One subject with RLS/WED (subject # 9) withdrew during the study due to personal time constraints and came for one week (baseline and 2 further treatments). Data from the remaining 12 subjects were used for all analyses, except when comparing with the control group. Only 11 subjects with RLS (7 men, 4 women) and 11 age- and sex-matched controls completed the study. See Table 1 for demographics.

 Table 1 Subject characteristics.

### Aim 1

RLS/WED severity decreased with 2 weeks of WBV as measured with IRLS; see Figure 1. The initial (baseline) IRLS average was 22.3 points (out of 40), indicating “severe RLS,” average IRLS after 2 weeks of vibration treatment: 17.1, indicating “moderate RLS.” Conversely, the initial IRLS average before sham treatment: 18.3, after 2 weeks of sham treatment: 17.7 (p = 0.6). The analysis for a possible carryover effect revealed a non-negligible carryover effect (p = 0.049), favoring the vibration treatment. After appropriately accounting for the cross-over design a 2-sample t-test was used to assess treatment effect.19 It was found to be significant with p = 0.013, 95% CI [−9.1, −1.3], see Table 2. RLS/WED severity decreased with 2 weeks of WBV as measured with RLS-6. The RLS-6 scales are typically used as a narrative context for the numeric values of the IRLS, and as such no formal analyses were intended or performed.27 When comparing the VAS answers for the RLS-6 questions from baseline (for those subjects in the “vibration first” group) or from the Friday before treatment (for those subjects in the “sham first” group) to the Friday at the end 2 weeks of treatment, questions 2, 1, and 5 demonstrated an improvement of over 2 points over the 2 weeks of treatment with an average decrease of 2.9, 2.5, and 2.1, respectively. None of the questions showed an increase in number on the VAS which would indicate exacerbation of symptoms. In the 2 weeks of sham treatment, however, the subjects indicated on average only a slight change, ranging from an increase of one point (question 4) to a decrease of 0.6 for questions 5 and 6.

 Table 2 Average IRLS changes over 2 weeks of treatment and sham.

### Aim 2

In order to explore Aim 2 we took several co-dependent steps and answered the following questions.

#### Step A

Can diminished SBF be possibly responsible for RLS/WED symptoms? To answer this question, we compared baseline SBF between subjects with RLS/WED with that of control subjects using a 2-sample t-test. Our assumption for equal variance of these 2 samples was confirmed with the Levene test. There was a difference in peripheral SBF in subjects with RLS/WED compared to control subjects, with an average flux of 25.5 (SD 2.78, n = 12) and 30.00 (SD 6.66, n = 11), respectively. The 2-sample t-test indicated that RLS subjects have on average a 4.5 unit lower SBF to the right foot (t-value = −2.16, df 11, p value = 0.04, 95% CI [−8.9, −0.2]).

#### Step B

Does SBF in subjects with RLS increase after WBV? To answer this question, we used a repeated measures design on all flux changes with WBV for each RLS/WED subject. The results showed an average increase of SBF of 7.8 units, 95% CI [5.9, 9.7], p < 0.001. There was no time effect, meaning that there was no change in pre- and post- SBF difference across the 2 weeks.

#### Step C

Does SBF increase more after WBV compared to sham treatment? Since the repeated measures showed that there was no time effect, we were justified in using simplified statistics by averaging all flux differences in order to answer this question. We confirmed that the WBV did indeed increase SBF compared to sham treatment: The blood flow change was significantly greater (p = 0.037) in the WBV than in the sham treatment. The WBV elicited an average flux change of 9.6, ranging from 3.05 to 18.8, 95% CI [0.5, 12.5], while the sham treatment's average change in SBF was 3.1, ranging from −0.8 to 4. The analysis for a possible carryover effect revealed a negligible carry-over effect (p = 0.28). After appropriately accounting for the crossover design, a 2-sample t-test was used to assess treatment effect.19

#### Step D

Does resting SBF after 2 weeks of WBV treatment increase? To answer this question, we compared RLS/WED subjects' resting SBF after 2 weeks of WBV treatment to their respective baseline SBF using a paired t-test. The results showed that 2 weeks of vibration treatment did not significantly increase resting blood flow in subjects with RLS/WED. The baseline flux for groups A and B, before their respective 2 weeks of WBV, was 24.4 (n = 11), compared to 26.1 after WBV treatment. The average change in flux was 0.45, t-value = 0.68, p value = 0.51.

## DISCUSSION

### Aim 1

The IRLS, a questionnaire reflecting RLS severity, showed a significant decrease in symptoms after the two-week treatment. The actual baseline IRLS scores were used as initial scores for all first interventions (vibration and sham), while the last score from the two intervention weeks was used as initial score for the next two weeks of intervention (sham and vibration). The established non-negligible carryover effect explains why the “baseline” scores for the sham treatment are lower (p = 0.01), with an average of 18.3 compared to the “baseline” scores for the vibration treatment (average 22.3). Hence, the shown treatment effect is more likely an underestimation of the actual treatment effect. The subjects undergoing 2 weeks of vibration experienced a decrease in RLS/WED symptoms, as evidenced in a decrease of over 5 points on the IRLS. In contrast, the subjects indicated no change in symptoms after 2 weeks of sham treatment (a decrease of 0.6 points, p = 0.6). According to a recent letter to the editor the minimal clinically significant change for the IRLS in clinical trials is a score of 3.28 Our findings of a decrease of over 5 points indicate that vibration has a positive effect on symptoms associated with RLS, leading to a clinically significant improvement. Burbank et al. have shown in several studies that vibrating pads, used at night, were able to improve sleep in patients with RLS.29,30 The authors believe that the mechanism of action of vibration lies in its ability to produce a counter stimulus to the brain to offset the sensations experienced with RLS/WED, called somatic hallucinations.29 While it might be possible that the subjects in our study experienced symptom relief based on sensory counter-stimulation through whole body vibration, we do not believe that this is the only reason. Flux is also altered in subjects with RLS/WED compared to matched controls,11 indicating that altered blood flow could very well be involved in the symptom etiology.

Three of the RLS-6 scales demonstrated a greater than 2-point improvement on the digital VAS. These particular 3 items establish that the subjects experienced decreased RLS/ WED symptom severity at night and during the day and that they were more satisfied with their sleep. None of the items increased on the VAS, showing that there was no exacerbation of symptoms. In the sham group, however, the subjects indicated on average only a slight, nonsignificant change, ranging from an increase of one point (question 4, symptom severity during day when resting) to a decrease of 0.6 for questions 5 and 6. The latter 2 items investigate RLS/WED symptom severity during the day when active and daytime tiredness.

### Aim 2

#### Step A

The evening peripheral SBF was significantly lower in subjects with RLS/WED compared to age and sex-matched controls. This does imply that altered SBF can theoretically explain symptoms associated with RLS/WED. This seems to be inconsistent with earlier findings,11 where subjects with RLS/WED were found to have higher baseline flux compared to controls. However, the data for that study were collected in the afternoon and recommendations were made to replicate the methods but conduct the study in the evening. That is what this current study was designed to do. Hence, our hypothesis, that, according to the circadian rhythm with which RLS/WED symptoms occur, peripheral blood flow decreases more in the evening in subjects with RLS/WED compared to normals, and thus elicit the sensations characteristic to RLS/WED, was supported. While our study only assessed skin blood flow, Larsson et al.9 assessed muscle oxygenation and capillary distribution in biopsies taken from the anterior tibialis in subjects with and without RLS/ WED. They found muscle fiber type and microvascularization differences between the two groups. The authors consequently hypothesized that those alterations could be the result of the body's attempt to counteract local hypoxia in the tissue.9

Peripheral blood flow is also affected by endogenous factors, such as nitric oxide (NO), a powerful vasodilator, induced by intra-arterial shear forces,31 and catecholamines, chemical messengers from the sympathetic nervous system.32 One study assessed NO production in subjects with RLS/WED after vibration,11 but we were not able to locate any studies measuring peripheral dopamine, a catecholamine, in subjects with RLS/ WED. Further studies are warranted in this direction.

#### Step B

The flux deficit that existed in RLS/WED subjects before the vibration intervention reversed when a vasodilating stimulus was given. The flux after WBV was as high in the RLS/WED group as in the controls, indicating that the vessels were able to dilate as much as necessary with vibration. In an earlier study, where data collection occurred in the early afternoon, flux also drastically increased with WBV, compared to baseline and compared to controls.11 This is an important finding because it indicates that a possible cause, hypoxia, can be treated with vibration and thus decrease symptoms associated with RLS/ WED. SBF in our control group did also increase significantly when comparing pre- to post-intervention data, and there was no difference in increase between the two groups.

#### Step C

WBV did increase SBF significantly more than did the sham treatment. Prior research11,14,21 has shown the circulation enhancing effects of WBV, and as such this result was expected. However, this study used the semi-squatting position, which is used for the WBV, as the sham treatment to ensure that the increase in blood flow was not the result of the position alone.

#### Step D

Our data did not show strong evidence of a correlation between change in IRLS score and change in blood flow (r = −0.2, p = 0.56), and thus we cannot support our notion that the change in blood flow over the 2-week treatment time explains the treatment effect found in the significant decline in RLS/ WED symptoms (as seen in decrease in IRLS scores). While after two weeks of WBV treatment symptoms associated with RLS were significantly lower compared to baseline, the flux did not exhibit corresponding significant changes. We expected the baseline flux (before WBV) to be higher by the end of the 2-week treatment sessions, but indeed, this was not the case. With these findings we can make two inferences: One, a 2-week treatment period with WBV is not long enough to make permanent changes in resting SBF, and two, it seems that WBV might be a treatment option, but it is uncertain if it is a cure. Further studies assessing long term benefits of WBV on RLS/WED symptomology are encouraged and warranted. We were, however, able to confirm that it was indeed the vibration that increased SBF and not the position that was assumed for the time of the treatment.11 This was not surprising as WBV is one of the treatments of choice when trying to increase blood flow. However, in order to validate our findings, we felt inclined to show the change.

### Limitations

Our study has some limitations. One, the placebo effect could have been stronger in the WBV group compared to the sham treatment group because the subjects were able to feel the difference. We tried to keep the effect as small as possible by explaining to the subjects that the position alone could have an effect on the outcome. Two, our treatment period of two weeks is very short. We justified this time because we have seen in previous studies12,24 that significant changes and most of the improvements can be seen in this short time period. Further research is necessary to explore possible continuous changes with longer treatments. Three, we did not conduct any follow-up assessment to evaluate effect maintenance. However, again, we feel that our findings encourage further research to answer the question on how long the achieved treatment effects last.

## CONCLUSION

A 2-week treatment with WBV decreases symptoms associated with RLS/WED as measured by the IRLS and RLS-6 questionnaires, but does not normalize skin blood flow in individuals with RLS/WED compared to healthy age-matched controls.

## DISCLOSURE STATEMENT

This was not an industry supported study. Financial support was provided by the Department of Exercise Sciences. The authors have indicated no financial conflicts of interest. This work was performed at the Human Performance Laboratory in the Exercise Sciences Department at Brigham Young University, Provo, Utah.

## ABBREVIATIONS

 IRLS International RLS Study Group Rating Scale LDI laser Doppler image QST Quantitative Sensory Testing RLS/WED restless legs syndrome/Willis Ekbom disease SBF skin blood flow WBV Vibration to the Whole Body

## REFERENCES

• 1 Barriere G, Cazalets J, Bioulac B, Tison F, Ghorayeb IThe restless legs syndrome. Progr Neurobiol; 2005;77:139-65.

• 2 Ekbom KRestless legs syndrome. Neurology; 1960;10:868-73, 13726241.

• 3 Ondo WEpidemiology of restless legs syndrome. Sleep Med; 2002;3:S13-5, 14592161.

• 4 Rajaram S, Shanahan J, Ash C, Walters A, Weisfogel GEnhanced external counter pulsation (EECP) as a novel treatment for restless legs syndrome (RLS): a preliminary test of the vascular neurologic hypothesis for RLS. Sleep Med; 2005;6:101-6, 15716213.

• 5 Lettieri C, Eliasson APneumatic compression devices are an effective therapy for restless legs syndrome: a prospective, randomized, double-blinded, sham-controlled trial. Chest; 2008;135:74-80, 19017878.

• 6 Eliasson A, Lettieri CSequential compression devices for treatment of restless legs syndrome. Medicine; 2007;86:317-23, 18004176.

• 7 Mitchell U, Myrer J, Johnson A, Hilton SRestless legs syndrome and near-infrared light: an alternative treatment option. Physiother Theory Pract; 2011;27:345-51, 20977377.

• 8 Wåhlin-Larsson B, Ulfberg J, Aulin KP, Kadi FThe expression of vascular endothelial growth factor in skeletal muscle of patients with sleep disorders. Muscle Nerve; 2009;40:556-61, 19623635.

• 9 Wåhlin-Larsson B, Kadi F, Ulfberg J, Aulin KPSkeletal muscle morphology in patients with restless legs syndrome. Eur Neurol; 2007;58:133-7, 17622717.

• 10 Salminen AV, Rimpila V, Polo OPeripheral hypoxia in restless legs syndrome (Willis-Ekbom disease). Neurology; 2014;82:1856-61, 24789861.

• 11 Mitchell U, Johnson PVibration and skin blood flow changes in subjects with restless legs syndrome. J Parkinsonism Restless Legs Syndrome; 2014;4:9-16.

• 12 Mitchell UUse of near-infrared light to reduce symptoms associated with restless legs syndrome: a case report. J Med Case Rep; 2010;4:286, 20731851.

• 13 Phillips B, Young T, Finn L, Asher K, Hening W, Purvis CEpidemiology of restless legs symptoms in adults. Arch Intern Med; 2000;160:2137-41, 10904456.

• 14 Lohman E, Petrofsky J, Maloney-Hinds C, Betts-Schwab H, Thorpe DThe effect of whole body vibration on lower extremity skin blood flow in normal subjects. Med Sci Monit; 2007;13:CR71-6, 17261985.

• 15 Zegarra-Parodi R, Park PY, Heath DM, Makin IR, Degenhardt BF, Roustit MAssessment of skin blood flow following spinal manual therapy: a systematic review. Man Ther; 2015;20:228-49, 25261088.

• 16 Mitchell UHMedical devices for restless legs syndrome - clinical utility of the Relaxis pad. Ther Clin Risk Manag; 2015;11:1789-94, 26664128.

• 17 International Restless Legs Study GroupIRLSSG diagnostic criteria for RLS (2012). Available at: www.irlssg.org.

• 18 Kohnen R, Benes H, Huelsmann J, Oertel W, Stiasny-Kolster K, Trenkwalder CSeverity rating in restless legs syndrome: validation of the RLS-6 scales. Sleep; 2004;27:A342(Abstract Suppl).

• 19 Wellek S, Blettner MOn the proper use of the crossover design in clinical trials: part 18 of a series on evaluation of scientific publications. Dtsch Arztebl Int; 2012;109:276-81, 22567063.

• 20 Shy ME, Frohman EM, So YTet al.Quantitative sensory testing: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology; 2003;60:898-904, 12654951.

• 21 Johnson PK, Feland JB, Johnson AW, Mack GW, Mitchell UHEffect of whole body vibration on skin blood flow and nitric oxide production. J Diabetes Sci Technol; 2014;8:889-94, 24876449.

• 22 Aubee APLaser doppler flowmetry; 2006Available at http://www.ele.uri.edu/courses/ele482/S06/Andrew_1.pdf.

• 23 Aukerman M, Aukerman D, Bayard M, Tudiver F, Thorp L, Bailey BExercise and restless legs syndrome: a randomized controlled trial. J Am Board Fam Med; 2006;19:487-93, 16951298.

• 24 Mitchell U, Johnson A, Myrer JComparison of two infrared devices in their effectiveness in reducing symptoms associated with RLS. Physiother Theor and Pract; 2010;27:352-59.

• 25 Bildt C, Alfredsson L, Punnett L, Theobald H, Torgen M, Wikman AEffects of drop out in a longitudinal study of musculoskeletal disorders. Occup Envir Med; 2001;58:194-9.

• 26 The International Restless Legs Syndrome Study GroupValidation of the International Restless Legs Syndrome Study Group rating scale for restless legs syndrome. Sleep Med; 2003;4:121-32, 14592342.

• 27 Allen R, Oertel W, Walters Aet al.Relation of the International Restless Legs Syndrome Study Group rating scale with the Clinical Global Impression severity scale, the restless legs syndrome 6-item questionnaire, and the restless legs syndrome-quality of life questionnaire. Sleep Med; 2013;14:1375-80, 24246378.

• 28 Allen RPMinimal clinically significant change for the International Restless Legs Syndrome Study Group rating scale in clinical trials is a score of 3. Sleep Med; 2013;14:1229, 24051118.

• 29 Burbank F, Buchfuhrer M, Kopjar BSleep improvement for restless legs syndrome patients. Part I: pooled analysis of two prospective, double-blind, sham-controlled, multi-center randomized clinical studies of the effects of vibrating pads on RLS symptoms. J Parkinsonism Restless Legs Syndrome; 2013;2013:1-10.

• 30 Burbank F, Buchfuhrer M, Kopjar BImproving sleep for patients with restless legs syndrome. Part II: meta-analysis of vibration therapy and drugs approved by the FDA for treatment of restless legs syndrome. J Parkinsonism Restless Legs Syndrome; 2013;3:11-22.

• 31 Ignarro L, Buga G, Wood K, Byrns REndothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A; 1987;84:9265-69, 2827174.

• 32 Taddei S, Pedrinelli R, Salvetti ASympathetic nervous system-dependent vasoconstriction in humans. Evidence for mechanistic role of endogenous purine compounds. Circulation; 1990;82:2061-7, 1978700.