Compiled by Ranile Samarakoon, founder of Stillr. Not a clinician. Stillr is developing a textile device for overnight wear, so we have a commercial interest in this field. That is exactly why this page states the limitations of the research as prominently as the results, cites every primary source, and makes no claim about any Stillr product. Nothing here is medical advice. If you have Restless Legs Syndrome, talk to your doctor.
Key findings
- Three published trials have tested compression modalities in Restless Legs Syndrome. Two were controlled, one was not.
- Pneumatic compression has the strongest evidence: a triple-blinded, sham-controlled randomised trial in 35 people found symptom severity fell from 14.1 to 8.4 on the 40-point IRLS scale, with one third of the treated group achieving complete symptom resolution and none in the sham group.
- Graduated compression stockings beat non-therapeutic placebo stockings in 63 pregnant women, a 58.3 per cent versus 41.9 per cent reduction in severity, p=0.018.
- Targeted foot pressure has never been tested against a sham. The only trial, in 30 people, was open-label with no control group, which makes its large reported effect very difficult to interpret.
- The placebo response in Restless Legs Syndrome is exceptionally high, averaging 6.58 points on the IRLS across 72 trials, more than double the 3-point threshold for clinical significance. This is the single biggest reason to distrust uncontrolled data in this condition.
- The 2024 AASM guideline did not assess compression at all. Its silence reflects an evidence gap, not evidence of absence.
What is Restless Legs Syndrome, and how is it measured?
Restless Legs Syndrome, also called Willis-Ekbom disease, is a sensorimotor disorder characterised by an urge to move the legs, usually with uncomfortable sensations, that worsens at rest and in the evening and is relieved by movement.
Severity in clinical trials is measured almost universally with the International Restless Legs Syndrome Study Group rating scale (IRLS), a 10-item patient-reported questionnaire scored 0 to 40. Scores of 1 to 10 are mild, 11 to 20 moderate, 21 to 30 severe, and 31 to 40 very severe. A change of 3 points is generally treated as the minimal clinically important difference.
Every number on this page is an IRLS score unless stated otherwise. That matters, because a headline like “17-point improvement” means nothing without knowing the scale runs to 40 and that placebo alone routinely moves it 6 to 9 points.
Does compression help restless legs syndrome?
There is limited but real evidence. Three published trials have tested compression modalities in Restless Legs Syndrome, and the two that used a control group both showed benefit over that control. None has been independently replicated.
| Intervention | Study | Design | n | IRLS result | Evidence quality |
|---|---|---|---|---|---|
| Pneumatic compression, lower leg, approx. 29 mmHg | Lettieri & Eliasson, CHEST, 2009 | Randomised, triple-blinded, sham-controlled | 35 | 14.1 to 8.4 in the treated group; p=0.006 between groups; one third achieved complete resolution versus zero on sham | Highest available. Rated moderate quality by the AHRQ 2012 review. Not replicated. |
| Graduated compression stockings, 23 to 32 mmHg, daytime wear | Kaplan et al., Rev Assoc Med Bras, 2024 | Randomised, placebo-stocking controlled, not blinded | 63 | 21.3 to 8.9 (58.3 per cent) versus 21.0 to 12.2 (41.9 per cent) on placebo; p=0.018 | Moderate. Participants could tell the stockings apart. Pregnancy-related RLS only. |
| Targeted foot pressure wrap, overnight | Kuhn et al., JAOA, 2016 | Single-arm, open-label, no control, no blinding | 30 | 25.1 to 7.8, a 17.2-point reduction; p<0.001 | Low. No sham, no randomisation, single centre, retrospectively registered, author institution received royalties. |
Pneumatic compression: the strongest trial in the field
The best-designed study of any compression modality in Restless Legs Syndrome is Lettieri and Eliasson (2009), published in CHEST and conducted at Walter Reed Army Medical Center. Thirty-five participants were randomised to a therapeutic device delivering about 29 mmHg of intermittent inflation, or to a visually identical sham delivering only 2 to 3 mmHg. Patients, investigators and referring physicians were all blinded. Devices were worn on both lower legs for at least one hour each evening for four weeks.
IRLS severity fell from 14.1 to 8.4 in the therapeutic group, significant against sham at p=0.006. Quality of life improved in social function (p=0.03), daytime function (p=0.02) and sleep quality (p=0.05). Sleepiness and fatigue both improved. One third of the therapeutic group achieved complete symptom resolution; none of the sham group did. No adverse events were reported.
The caveats: 35 people is a small trial, it has never been replicated, and two later registered trials of other pneumatic devices (NCT03852771, NCT01861652) have published no results at all. Absence of published results from registered trials is itself a signal worth noting.
Compression stockings in pregnancy
The only randomised trial of compression stockings in Restless Legs Syndrome is Kaplan and colleagues (2024), which enrolled 63 pregnant women at 27 to 37 weeks gestation. Thirty-two received graduated compression stockings at 23 to 32 mmHg; 31 received 100-denier non-therapeutic placebo stockings. Both groups wore them from waking until bedtime for three weeks.
Severity fell 58.3 per cent in the compression group versus 41.9 per cent on placebo, significant at p=0.018, and benefit began significantly sooner (3.93 versus 5.09 days, p=0.007). But sleep quality and general wellbeing showed no significant between-group difference, the placebo group improved substantially on its own, and the trial was not blinded because participants could feel which stockings were therapeutic. Pregnancy-related Restless Legs Syndrome may also not generalise to primary RLS.
Targeted foot pressure: a large effect with no control group
The only published clinical study of targeted foot pressure for Restless Legs Syndrome is Kuhn, Olson and Sullivan (2016) in the Journal of the American Osteopathic Association. Forty-seven patients with moderate to severe primary RLS were enrolled across two offices in Erie, Pennsylvania between 2009 and 2012; 30 completed the eight-week protocol.
The reported effect is large. Mean IRLS fell from 25.05 to 7.83, a 17.22-point reduction (95 per cent CI 14.92 to 19.52, p<0.001). Nightly sleep loss fell from 119.5 to 22.1 minutes. Ninety per cent of patients rated themselves much or very much improved.
The limitations are fundamental rather than incidental. There was no sham device, no blinding, no control group and no randomisation. The sample was 30 people at a single centre. Registration on ClinicalTrials.gov occurred in 2015, three years after the study finished. The lead author’s institution subsequently received royalties from the device rights, and the inventor was acknowledged as a financial supporter. The authors’ comparison against ropinirole was made against published data from three earlier, separate drug trials rather than a concurrent comparison arm, so it is not a head-to-head result. No independent replication has ever been published.
The device was granted FDA De Novo classification in December 2013 as a Class I device, the lowest risk category. The FDA’s own decision summary is candid: it noted the placebo response on the IRLS “has been reported as greater than 40 percent in randomized, controlled trials,” and that “as the supporting data were derived from a single arm, historically controlled trial, there is some uncertainty regarding the benefits.” Class I clearance is a safety determination, not proof that something works.
Why is the placebo response such a problem in this research?
Because it is among the highest in medicine. A 2017 meta-analysis in Neurology pooling 72 trials found a mean placebo improvement of 6.58 points on the IRLS, more than double the 3-point minimal clinically important difference.
In the 2014 New England Journal of Medicine pregabalin trial, 46.8 per cent of placebo recipients rated themselves much or very much improved at 12 weeks. Placebo responder rates consistently sit in the 40 to 47 per cent range across large randomised trials.
Applied to the uncontrolled 17.22-point foot pressure result: subtracting a typical placebo response of roughly 6.6 to 8.9 points leaves an estimated residual of about 8 to 10 points. That would still be clinically meaningful, but it is roughly half the headline number, and it is an estimate borrowed from other trials rather than a measurement. This is why an unblinded study in this condition cannot tell you what an intervention does, no matter how large its reported effect.
What changed in the 2024 treatment guidelines?
The American Academy of Sleep Medicine reversed its position on dopamine agonists. Drugs that carried strong recommendations in 2012 now carry conditional recommendations against their use.
| Agent or intervention | 2012 position | 2024 position |
|---|---|---|
| Pramipexole, ropinirole | Strong recommendation for | Conditional recommendation against |
| Rotigotine, levodopa | Recommended | Conditional recommendation against |
| Cabergoline | Recommended | Strong recommendation against, fibrotic complications |
| Gabapentin enacarbil, gabapentin, pregabalin | Secondary | Strong recommendation for, first line |
| Intravenous iron | Not addressed | Strong recommendation for, where ferritin is below 100 micrograms per litre or transferrin saturation below 20 per cent |
| Bilateral high-frequency peroneal nerve stimulation | Not available | Conditional recommendation for |
| Compression of any kind | Not addressed | Not addressed, outside the scope of the systematic review |
The reason for the reversal is augmentation: a progressive, treatment-caused worsening of Restless Legs Syndrome, where symptoms start earlier in the day, spread to other body parts, and intensify. It accumulates at roughly 7 to 10 per cent per year, reaching an estimated 20 to 30 per cent at three years and 35 to 50 per cent at five years. The 2014 NEJM trial of 719 patients measured augmentation at 1.7 per cent for pregabalin versus 9.0 per cent for pramipexole at 52 weeks. Our guide to Restless Legs augmentation covers the warning signs.
What non-drug options have the best evidence?
Bilateral high-frequency peroneal nerve stimulation has the strongest evidence of any non-drug device, and is the only device to receive a recommendation in the 2024 AASM guideline. In a multisite randomised sham-controlled crossover trial (n=37), IRLS reduction was 6.81 points on active treatment versus 3.38 on sham (p<0.01), with a 66 per cent clinically significant responder rate against 17 per cent for sham. A larger pivotal trial of 133 patients met all seven primary and secondary endpoints.
Beyond devices, the 2019 systematic review by Harrison and colleagues found that compression devices, repetitive transcranial magnetic stimulation, exercise and acupuncture may reduce severity, while vibration pads and cryotherapy may improve sleep without reducing core symptoms. A vibratory counter-stimulation pad cleared in 2013 significantly improved sleep quality across two pooled trials (n=158) but did not significantly reduce IRLS severity versus sham. See our wider guide to drug-free Restless Legs treatment.
Should you wear compression to bed?
For venous conditions, standard clinical guidance is no. Graduated compression works with gravity while you are upright, so there is no venous benefit while lying down, and there are real risks: arterial compromise when supine, skin breakdown, and pressure-point injury. Overnight compression warrants particular caution with diabetes, peripheral neuropathy, peripheral arterial disease, or any condition affecting sensation or circulation in the feet.
Devices designed for overnight counter-stimulation are a different proposition mechanically, because the intended action is sensory input rather than venous return. But that difference cuts both ways: it means venous guidance does not automatically apply, and it also means venous safety data does not automatically transfer. On pressure specifically, therapeutic compression in the 15 to 40 mmHg range corresponds to roughly 2 to 5.3 kPa, while measured pressure pain thresholds on the medial arch of the foot sit around 273 to 500 kPa, roughly one hundred times higher. Comfort at these pressures is plausible on that basis, but plausibility is not a safety study.
Ask your doctor before wearing any compression garment overnight.
Why might pressure on the foot affect a neurological condition at all?
The proposed mechanism is spinal gate control: large-diameter sensory nerve fibres carrying touch and pressure signals can suppress transmission of other sensory signals at the level of the spinal cord.
Gate control theory, proposed by Melzack and Wall in 1965, holds that large myelinated A-beta fibres from skin mechanoreceptors synapse onto inhibitory interneurons in the dorsal horn, dampening other sensory transmission. The sole of the foot is unusually well suited to engaging this. Microneurography work by Kennedy and Inglis recorded 104 cutaneous mechanoreceptors and estimated roughly 4,000 tactile afferents per foot sole.
For sustained pressure rather than transient touch, the relevant receptors are the slowly adapting types: Merkel disc afferents fire throughout the entire duration of a maintained stimulus, and Ruffini endings maintain discharge during sustained skin stretch. A constant-pressure garment therefore generates continuous afferent signalling, not a single burst.
There is a plausible dopaminergic link as well. Restless Legs pathophysiology centres on the A11 diencephalospinal dopamine system, the only source of spinal dopamine. Animal work has shown that A11-homologous neurons release dopamine in response to tactile stimuli, and human PET imaging has shown that peripheral electrical stimulation causes measurable dopamine release in the striatum and nucleus accumbens.
None of this is evidence that a given device works. Mechanistic plausibility explains why a hypothesis is worth testing. It does not substitute for the test. The honest summary of this field is that the biology is interesting and the trial evidence is thin.
What would actually settle the question?
A properly blinded, sham-controlled randomised trial of targeted foot pressure, powered to detect an effect above the roughly 6.6-point placebo baseline that characterises Restless Legs research. No such trial has ever been conducted.
That is the single most useful sentence on this page. Everything else is a description of what we have in the absence of it.
Sources
- Lettieri CJ, Eliasson AH. Pneumatic compression devices are an effective therapy for restless legs syndrome: a prospective, randomized, double-blinded, sham-controlled trial. CHEST. 2009. PMID: 19017878. doi:10.1378/chest.08-1665. NCT00479531.
- Eliasson AH, Lettieri CJ. Sequential compression devices for treatment of restless legs syndrome. Medicine. 2007. doi:10.1097/MD.0b013e31815b1319.
- Kaplan O, Baser M, Ozgun MT. Effect of compression stockings on restless legs syndrome in pregnancy. Rev Assoc Med Bras. 2024. PMID: 39166663. doi:10.1590/1806-9282.20240145. NCT05795868.
- Kuhn PJ, Olson DJ, Sullivan JP. Targeted pressure for moderate to severe primary restless legs syndrome. J Am Osteopath Assoc. 2016. PMID: 27367949. doi:10.7556/jaoa.2016.088. NCT02386423 (registered retrospectively, 2015).
- US Food and Drug Administration. De Novo classification decision summary, DEN110009, 18 December 2013. Product code OTX.
- Winkelman JW, et al. Treatment of restless legs syndrome and periodic limb movement disorder: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2024. PMID: 39324694. doi:10.5664/jcsm.11390.
- Allen RP, et al. Comparison of pregabalin with pramipexole for restless legs syndrome. N Engl J Med. 2014;370:621-631.
- Silva MA, et al. Placebo and nocebo responses in restless legs syndrome: a systematic review and meta-analysis. Neurology. 2017. PMID: 28490647.
- Charlesworth JD, et al. Bilateral high-frequency non-invasive peroneal nerve stimulation for restless legs syndrome: a randomized sham-controlled crossover trial. J Clin Sleep Med. 2021. PMC8656897.
- Harrison EG, Keating JL, Morgan PE. Non-pharmacological interventions for restless legs syndrome: a systematic review of randomised controlled trials. Disabil Rehabil. 2019. PMID: 29561180.
- Agency for Healthcare Research and Quality. Comparative effectiveness review: treatment for restless legs syndrome. 2012.
- Melzack R, Wall PD. Pain mechanisms: a new theory. Science. 1965;150:971-979.
- Kennedy PM, Inglis JT. Distribution and behaviour of glabrous cutaneous receptors in the human foot sole. J Physiol. 2002. PMC2290100.
- Garcia-Borreguero D, et al. IRLSSG Task Force report on placebo response in RLS clinical trials. Sleep Med Rev. 2024.