Nine hundred patients meeting International Restless Legs Syndrome Study Group criteria for RLS were examined and treated by the author between 2004 and 2014. Of full respondents, up to 51% had the full triad of RLS, migraine headaches, and bruxism. When treated with dopamine agonists, 80% responded for symptoms of RLS and up to 70% for RLS-related bruxism.

Restless limbs syndrome (RLS) or Willis-Ekbom disease (WED) was described by Sir Thomas Willis in 16851 and was brilliantly defined by Karl-Axel Ekbom in 1945.2 Ekbom himself estimated a prevalence of 5% to 10% of the Scandinavian population. However, subsequent authors have suggested that as many as 29% of the population may have RLS. For argument’s sake, many use 10% as a useful estimate, affecting roughly 35 million Americans.3 I should note here that I am intentionally using “RLS” to stand for restless “limbs” syndrome, not the more common name of restless “legs” syndrome, for reasons that will become clear in this article.

The diagnosis of RLS is made based on four criteria proposed by the International Restless Legs Syndrome Study Group (IRLSSG).

  • An urge to move the legs usually but not always accompanied by or felt to be caused by uncomfortable and unpleasant sensations in the legs.
  • The urge to move the legs and any accompanying unpleasant sensations begin or worsen during periods of rest or inactivity such as lying down or sitting.
  • The urge to move the legs and any accompanying unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues.
  • The urge to move the legs and any accompanying unpleasant sensations during rest or inactivity only occur or are worse in the evening or night than during the day.
  • The occurrence of the above features is not solely accounted for as symptoms primarily due to another medical or behavioral condition.4

RLS is divided into primary RLS, which is a heritable form (predominantly autosomal dominant), and secondary RLS. Secondary RLS has been reported in association with iron deficiency, renal failure, hemodialysis, neuropathy, pregnancy, multiple sclerosis, psychiatric conditions, and Parkinson’s disease.5-9 RLS can be a side effect of medications, especially neuroleptics, antihistamines, antidepressants, and metoclopramide.10

In spite of the high prevalence of the disease and widespread appreciation of the diagnostic criteria, RLS remains underdiagnosed and undertreated, resulting in substantial impact on the quality of life of patients.

The Journey

My journey into the world of RLS began with an interest in movement disorders and Parkinson’s disease that was nurtured during my residency under Melvin Yahr, MD. During my fellowship training in neuromuscular neurology and electrodiagnostic medicine, I became fascinated by the search for treatable neuropathies. Electromyographers have a privilege not shared by many other specialists—we stare at feet for long periods of time. From time to time, I became aware that patients with complaints of numbness, tingling, and burning pains in their feet may have accompanied, stereotyped movements of the feet and toes. These are asynchronous, arrhythmic, writhing, and wiggling movements of varying amplitudes and frequencies. I noticed that, like the rest tremor of Parkinson’s disease, the patients could suppress the toe movements when the movements were pointed out. However, once the patient became distracted, the movements would begin again. I soon made the connection that patients with these movements and “classical” neuropathic complaints frequently had perfectly normal electrodiagnostic studies. Many of these patients had clinical features of RLS that I thought might explain their neuropathic symptoms.

I became increasingly aware that my RLS patients frequently suffered from headaches and my migraine patients frequently suffered from insomnia and RLS. The literature strongly supports the association of RLS and headaches, specifically migraines.11-15 Gupta et al reported headaches in 51% of RLS patients with migraines in 44%.11 The association remains when looking at RLS frequency in headache patients.12,15,16

Around 2007, one of my patients volunteered that her chronic bruxism had resolved with treatment of her RLS. Lavigne and Montplasir first reported an association between RLS and bruxism,17 an observation later confirmed by Ahlberg et al.18 Both bruxism and the periodic limb movements of sleep (PLMS) that are associated with RLS are movement disorders and parasomnias and occur in early stages (stages 1 and 2) of sleep. Both have high prevalence affecting up to 20% of the adult population.19 Bruxism, and associated temporomandibular disorders,  also have been associated with headaches in adults and children20-23 and insomnia.24

The Study

We created a 33-question survey that addresses age, location of symptoms, and diurnal variation (both at onset of symptoms and at the time of diagnosis) as well as personal symptoms, associated headaches, bruxism, family history, and response to treatment. Charts were reviewed and interviews conducted to complete and clarify the data. The questionnaire is available on the Willis-Ekbom Disease Foundation website (

Approximately 600 questionnaires were obtained retrospectively from patients who had been treated in the practice from 2004 to 2012. About 400 more were obtained, prospectively, from patients who were seen in the practice from 2012 to 2014. The first group of patients met IRLSSG criteria for RLS and predominantly presented with neuropathic complaints of pain, paresthesias, and dysesthesias. Many of the subsequent 400 patients were referred with headaches but had clear criteria for RLS upon system review. What was remarkable was that full data analysis at 600 and 900 patients showed precisely the same demographic characteristics suggesting that the populations were similar whether they were selected based on neuropathic complaints or presentation with primary headache.

The Results


Figure 1

The results of the study are analyzed in two ways. First, statistics are given based on the entire population of 900 patients. However, not all patients answered all questions. A second analysis is done based on “full respondents” who gave answers to all questions involved in determining a specific data point. Seventy-four percent of the population were female. The mean age of symptom onset was 40 years and at diagnosis, 53 years. Surprising numbers of patients had symptoms in the arms and these could often be the presenting location. In fact, symptoms began in the legs (60%), in the arms (13%), or in both arms and legs (27%). At the time of diagnosis, 13 years later, the most common presentation was with symptoms in arms and legs (57%), legs (36%), and arms alone (7%). (See Figure 1.)


Figure 2

Principal symptoms included the following: insomnia (81%), headaches (75%), pain (72%), numbness and tingling (70% and 66%, respectively), cramp-like pain (37%), bruxism (35%), and aching (32%). Pain complaints could overlap as any individual patient could have multiple symptoms. Dysesthetic pains included the following: burning (25%), throbbing (17%), electricity (15%), coldness (13%), crawling (13%), pulling (11%), and flowing (4%).25 (See Figure 2.)

Pain and/or stiffness upon awakening in the morning was reported by 52.6%, and 41.8% reported similar complaints during the day with sitting. These complaints did improve with movement, but not immediately upon initiation of movement. Rather, morning pain and stiffness required a mean of 25.6 minutes (mode = 10 minutes) to improve and daytime pain and stiffness after sitting improved with a mean latency of 20.2 minutes (mode = 5 minutes).26

Seventy-five percent of the entire RLS group suffered from headaches. These had clinical features of migraines including photophobia (86%), phonophobia (73%), nausea (66%), and vomiting (24%); 47% of respondents reported a positive family history for headaches. In addition, 43% reported a family history of chronic pain, 25% a family history of insomnia, and 15% a family history of bruxism (usually in patients’ children). Over a quarter (27%) gave a personal history and 15% gave a family history of growing pains.

Sixty-one percent of full respondents reported having bruxism. Forty percent of all patients and 70% of full respondents reported improvement of the bruxism with dopamine agonists. A review of 75 randomly selected charts showed that the mean effective dose of pramipexole (n=53) was 0.98 mg per day (range 0.125 mg – 4 mg/day) and of ropinirole (n=14) was 0.32 mg day (range 0.5 mg – 10 mg/day).27


Figure 3

Of 470 patients who responded fully to all questions regarding headaches and bruxism, 391 (78%) had migraines; 304 (61%) had bruxism; and 264 (52%) had the full triad (see Figure 3).28

Of importance to note is that during the period between onset of symptoms and diagnosis, patients are searching for answers and therapies. Several hundred (236) patients reported having a total of 262 surgical procedures often designed to treat their RLS-related symptoms. The surgeries included: knee or other joint surgeries (212), back surgery (65), carpal tunnel or other nerve surgery (43), neck surgery (13), and other surgery (29). Of these 236 patients, only 34% reported a lasting benefit from their surgical intervention.28

Discussion—Incorporating Real Patient Experience

With more data in hand than I had imagined, I revisited my initial question: Why is such a common condition as RLS so under-recognized and undertreated? I also now knew about the significant delay between the onset of symptoms at age 40 and diagnosis at age 53.

I began by reviewing the IRLSSG diagnostic criteria and comparing them to what my patients were telling me about RLS:

“Uncomfortable and unpleasant sensations”: In 10 years, I cannot recall hearing a patient offer these complaints. Certainly, patients are uncomfortable and their sensations are unpleasant. However, this requires some translation on the part of the physician. I return to Sir Thomas Willis: “When being in bed they betake themselves to sleep, presently in the arms and legs, leaping and contractions of the tendons and so great a restlessness and tossing of the members ensue, that the diseased are no more able to sleep, than if they were in the place of the greatest torture.”1 Although written more than three centuries ago, I cannot imagine a better description of RLS. It includes symptoms of pain, insomnia, restlessness, and probably even PLMS. It suggests that these “tortured” patients have degrees of pain that are not advised by the current diagnostic criteria. Pain complaints in our patients (72%) include burning, throbbing, electric shocks, hot and cold sensations, crawling, aching, and other types. The descriptions of pain are as numerous as the patients themselves. Each experience is unique and individualized. The great majority also complain of paresthesias (numbness in 70% and tingling in 66%). A greater appreciation of the “real patient experience” of RLS might make diagnostic recognition easier.25

“In the legs”: This part of the definition may be deceptive and misleading to many. In fact, my patients report a high percentage of arm involvement at the time of symptom onset (40.5%) and even more at the time of diagnosis (64%). What is even less appreciated is that symptoms can begin in the arms (13.4%) and can even present as a hemi-syndrome involving the ipsilateral arm and leg (7%). In fact, by the time the diagnosis is made (mean age: 53), the single most common presenting location of symptoms involves both arm and leg (57.1%). This is the main reason why I favor the name “restless limbs syndrome.”

Willis-Ekbom disease is historically accurate but not descriptive. Restless legs syndrome is historically important but diagnostically misleading. Appreciation of the arm involvement in RLS also unravels other diagnostic puzzles. We presented 40 patients referred to our electromyography (EMG) laboratory with classical symptoms of carpal tunnel syndrome including numbness, tingling, and pain in one or both arms, worse at night, interrupting sleep, and improvement with movement and shaking. All 40 patients had normal electrodiagnostic studies; 26 of 27 had improvement or resolution of their symptoms with dopamine agonists. (The other 13 did not choose and/or require treatment.)29

“Partially or totally relieved by movement”: This statement is true, but it is also somewhat misleading. Patients report improvement with movement but often describe a significant latency required for that improvement to occur. More than half (52%) describe pain and stiffness in the morning upon awakening, which can require more than an hour to improve. In fact, some patients’ most severe symptomatic period is in the morning upon awakening, and rare patients only voice complaints at this time. Similarly, 42% of patients note the development of pain and stiffness with prolonged sitting or inactivity during the day. This also can require a significant latency for improvement.26 Without this recognition, one might preclude the diagnosis of RLS if patients do not immediately improve with movement.

“Urge to move”: No one complains of this either, although many will describe a “need to move.”30 It truly behooves the clinician to pursue this history with questions like: “What makes the symptoms better?” or “Are your complaints worse at rest or with activity?” Patients will volunteer that they get out of bed and walk at night, or shift their weight from side-to-side with prolonged sitting. When trying to understand what my patients were experiencing, I was reminded of an older concept once taught in psychiatry: akathisia. Akathisia (from the Greek and literally meaning “inability to sit”) is a state of agitation, distress, and restlessness that is classically a side effect of antipsychotic drugs (dopamine antagonists) but, curiously, also can be seen, like RLS, as a side effect of antidepressant drugs. Akathisia can also be associated with severe pain, particularly around the knees.

“Worse in the evening or night than during the day”: The diurnal variability of RLS is a classic hallmark of the condition but may be misleading when patients have severe daytime symptoms. In our cohort, 50% had symptom onset in the evening or at night. However, 13 years after onset, at the time of diagnosis, a full 74% reported daytime symptoms. These interfere with productivity on the job as sufferers are unable to sit and perform their duties. Commuting in rush hour traffic is a misery for many. The presence of significant daytime symptoms also has important ramifications regarding treatment. When symptoms occur only in the evening and night, a single dose of a dopamine agonist after dinner may suffice. Alternatively, nonpharmacological treatments like the FDA-approved Relaxis device can be used alone or added to drug therapies. However, when prominent daytime symptoms are present, long-acting dopamine agonists or split doses during the day are required.

Conspicuously absent from the diagnostic criteria is, perhaps, the most important nocturnal symptom of RLS: insomnia. Our work suggests that it is the most common symptom as well, affecting 81% of patients. Estimates of the number of Americans suffering from insomnia range from 30 to 90 million. Applying simple math (81% of 35 million RLS sufferers) makes the point that 30% or more of insomnia in the United States may be directly attributable to RLS! Yet, how many of us consider treating insomnia with dopamine agonists? Perhaps we should when RLS is the cause.31

Not solely accounted for as symptoms primarily due to another…condition“: This is the idea of excluding mimicking conditions and leads to the important issue of RLS and comorbidities. When a disease affects 10% of a population, there will be considerable overlap with other conditions. However, it may be more important to consider RLS than to exclude it. For example, we have been taught that cramps exclude the diagnosis of RLS. True electrophysiological cramps are probably not associated with RLS but you cannot tell patients what they are experiencing. A full 37% of patients complain of cramp-like pains, usually at night. Patients with diabetes, lupus, arthritis, and other common diseases have primary RLS. All too often, their RLS symptoms are quickly ascribed to their other medical conditions and safe and appropriate treatments for RLS are overlooked.

As a case in point, we analyzed data on 426 patients with RLS who underwent upper extremity EMG studies; 316 (74.2%) had normal EMGs and 110 (25.8%) were positive, most often for carpal tunnel syndrome (38% mild, 38% moderate, and 23% severe). Ninety percent (n=62) of the EMG+ patients and 91% of the EMG- patients reported a favorable response to dopamine agonists. All 6 of 6 patients who underwent carpal tunnel release did not improve. We have had a similar experience with RLS patients with diabetic neuropathy, fibromyalgia, and pain with connective tissue diseases. Many have responded to dopamine agonists after failing therapies for their other diagnoses. Perhaps it is more important to be inclusive when it comes to RLS rather than excluding the diagnosis and treatment in favor of treating the comorbid condition.


Disease recognition of primary RLS would be improved by incorporating real patient experience into the definition:


David J. Dickoff, MD

Primary RLS is a common inherited disease characterized by pain, paresthesias, and dysesthesias in the arms and legs associated with akathisia. Symptoms occur at rest and inactivity and begin to improve with movement. RLS is frequently associated with migraine headaches and insomnia and may be a major cause of bruxism (“restless jaw”). RLS can mimic many other conditions and should be considered in their differential diagnoses. RLS is a progressive disease with symptoms spreading (anatomically) over space and time (nocturnal to diurnal). The diagnosis of RLS is supported by family history of RLS or related conditions and by response to an appropriate trial of dopamine agonists.

Our cohort study also suggests various avenues of future research. The apparent association of migraine headaches and RLS raises a possible genetic link between the two conditions, perhaps contributing to a channelopathy causing centrally mediated pain. One of my patients went so far as to describe his throbbing nocturnal dysesthesias as “migraines in the legs.”

The work also suggests that RLS may be the missing link connecting bruxism and temporomandibular disorders with headaches and that the headaches should be treated as migraines. If a double-blind, prospective study confirms the responsiveness of bruxism to dopamine agonists, it would offer the first pharmacological treatment for this common condition that causes billions of dollars of dental and periodontal morbidity. Improved recognition of the RLS-migraine-bruxism triad may allow for the treatment of chronic insomnia in millions of Americans and shorten the time between onset of symptoms and diagnosis during which patients may be exposed to inappropriate therapies and surgeries.

David J. Dickoff, MD, is a consulting neurologist in Yonkers, NY. He is a clinical assistant professor at Mount Sinai Medical Center in New York and is the medical director of the Riverside Parkinson’s Self-Support Group in Yonkers. Questions and comments can be addressed to the author at Metropolitan Neurological Consultants, PC, 984 N Broadway, Suite 509, Yonkers, NY 10701.


1. Willis T. The London Practice of Physick, Or The Whole Practical Part of Physick. London: Passet and Crooke; 1685.

2. Ekbom KA. Restless legs syndrome. Acta Med Scand. 1945;158(suppl):4-122.

3. Yeh P, Walters AS, Tsuang JW. Restless legs syndrome: a comprehensive overview on its epidemiology, risk factors, and treatment. Sleep Breath. 2012;16(4):987-1007. doi: 10.1007/s11325-011-0606-x.

4. International Restless Leg Syndrome Study Group, revised September 24, 2013 from

5. Lecendreux M. S24. A psychiatric co-morbidity in restless legs syndrome and periodic limb movements in sleep. Sleep Med. 2007;8:S28. doi: 10.1016/S1389-9457(07)70101-6.

6. Allen RP, Picchietti DL, Garcia-Borreguero D, et al. Restless legs syndrome/Willis-Ekbom disease diagnostic criteria: updated International Restless Legs Syndrome Study Group (IRLSSG) consensus criteria—history, rationale, description, and significance. Sleep Med. 2014;15(8):860-873. doi: 10.1016/j.sleep.2014.03.025.

7. Zhu XY, Liu Y, Zhang XJ, et al. Clinical characteristics of leg restlessness in Parkinson’s disease compared with idiopathic Restless Legs Syndrome. J Neurol Sci. 2015 Jul 9. [Epub ahead of print]

8. Sieminski M, Losy J, Partinen M. Restless legs syndrome in multiple sclerosis. Sleep Med Rev. 2015;22:1522. doi: 10.1016/j.smriv.2014.10.002.

9. Bastia JK, Bhoi SK, Kalita J, Misra UK. Neuropathy in a cohort of restless leg syndrome patients. J Clin Neurosci. 2015;22(8):1314-1318. Doi: 10.1016/j.jocn.2015.01.032.

10. Bliwise DL, Zhang RH, Kutner NG. Medications associated with restless legs syndrome: a case-control study in the US Renal Data System (USRDS). Sleep Med. 2014;15(10):1241-1245. doi:10.1016/j.sleep.2014.05.011.

11. Gupta R, Lahan V, Goel D. Primary headaches in restless legs syndrome patients. Ann Indian Acad Neurol. 2012;15(Suppl 1):S104-S108. doi: 10.4103/0972-2327.100031.

12. Rhode AM, Hösing VG, Happe S, Biehl K, Young P, Evers S. Comorbidity of migraine and restless legs syndrome—a case-control study. Cephalalgia. 2007;27(11):1255-1260.

13. Cologno D, Cicarelli G, Petretta V, d’Onofrio F, Bussone G. High prevalence of Dopaminergic Premonitory Symptoms in migraine patients with Restless Legs Syndrome: A pathogenetic link? Neurol Sci. 2008;29 Suppl 1:S166-8. doi: 10.1007s10072-008-0915-4.

14. Chen P-K, Fuh J-L, Chen S-P, Wang S-J. Association between restless legs syndrome and migraine. J Neurol Neurosurg Psychiatry. 2010;81(5):524-528. doi: 10.1136/ jnnp.2009.200030.

15. Cannon PR, Larner AJ, Article R. Migraine and restless legs syndrome: is there an association? J Headache Pain. 2011;12(4):405-409. doi: 10.1007/s10194-011-0357-x.

16. D’Onofrio F, Bussone G, Cologno D, et al. Restless legs syndrome and primary headaches: a clinical study. Neurol Sci. 2008;29 Suppl 1:S169-172. Doi: 10.1007/s10072-008-0916-3.

17. Lavigne GJ, Montplaisir JY. Restless legs syndrome and sleep bruxism: prevalence and association among Canadians. Sleep. 1994;17(8):739-743.

18. Ahlberg K, Ahlberg J, Könönen M, Partinen M, Hublin C, Savolainen A. Reported bruxism and restless legs syndrome in media personnel with or without irregular shift work. Acta Odontol Scand. 2005;63(2):94-98. doi: 10.1080/000163503100006753.

19. Lavigne GJ, Khoury S, Abe S, Yamaguchi T, Raphael K. Bruxism physiology and pathology: An overview for clinicians. J Oral Rehabil. 2008;35(7):476-494. doi:10.1111/j.1365-2842.2008.01881.x.

20. Carra MC, Bruni O, Huynh N. Topical review: sleep bruxism, headaches, and sleep-disordered breathing in children and adolescents. J Orofac Pain. 2012;26(4);267-276.

21. Fernandes G, Franco AL, Gonçalves DA, Speciali JG, Bigal ME, Camparis CM. Temporomandibular disorders, sleep bruxism, and primary headaches are mutually associated. J Orofac Pain. 2013;27(1):14-20. doi: 10.11607/jop.921.

22. Bruni O, Fabrizi P, Ottaviano S, Cortesi F, Giannotti F, Guidetti V. Prevalence of sleep disorders in childhood and adolescence with headache: a case-control study. Cephalalgia. 1997;17(4):492-498.

23. Vendrame M, Kaleyias J, Valencia I, Legido A, Kothare SV. Polysomnographic findings in children with headaches. Pediatr Neurol. 2008;39(1):6-11. doi: 10.1016/j.pediatrneurol. 2008.03.007.

24. Bailey DR. Tension headache and bruxism in the sleep disordered patient. Cranio. 1990;8(2):174-182.

25. Dickoff DJ, Serkin R, Dontje S. Restless limbs syndrome (RLS): The real patient experience. Abstract presented as poster at: annual meeting of the American Neurological Association; September 2014; Baltimore.

26. Dickoff DJ, Serkin R, Dontje S. Pain and stiffness on initiating movement in restless limbs syndrome (RLS): An important diagnostic consideration. Abstract presented at: annual meeting of the American Academy of Neurology; April 2015; Washington, DC.

27. Dickoff DJ, Sirkin R, Dontje S, Ditkoff N, Silberman N. Restless limbs syndrome (RLS) and bruxism: A subgroup analysis and dose-response data. Abstract presented at: annual meeting of the American Academy of Neurology; April 2013; San Diego.

28. Dickoff DJ, Serkin R, Dontje S. Restless limbs syndrome (RLS), migraine syndrome and bruxism: an important clinical triad. Presented as platform and poster at: annual meeting of the American Neurological Association; September 2013; New Orleans.

29. Dickoff DJ, Dontje S. Restless limbs syndrome (RLS) mimicking carpal tunnel syndrome. Abstract presented at: annual meeting of the American Association of Neuromuscular Diseases and Electrodiagnostic Medicine; October 2012; Orlando, Fla.

30. Kerr S, McKinon W, Bentley A. Descriptors of restless legs syndrome sensations. Sleep Med. 2012;13(4):409-413. doi: 10.1016/j.sleep.2011.11.020.

31. Dickoff DJ, Sirkin R, Dontje S. Restless limbs syndrome (RLS), bruxism, and migraine triad: An important diagnostic consideration in the diagnosis and treatment of insomnia. Abstract presented as poster at: annual meeting of the American Society of Sleep Medicine; June 2015; Seattle.