Restless legs syndrome (RLS) was originally described by Willis in 1685 (1) and was then more fully medically characterized by Ekbom in 1950 (2). RLS is a major focus in the sleep clinic because it negatively impacts sleep, quality of life, and overall health (3, 4). The current diagnostic criteria for RLS include the urge to move accompanied by uncomfortable sensations during periods of rest that worsen in the evening (5). The attributes of the diagnostic criteria come from the Montreal Group, where they studied the suggested immobilization test. During an hour of voluntary immobilization, the subject is observed for periodic leg movements. Often the symptom intensity increases with increased duration of immobilization (6). These symptoms are often described by the subject as feeling, “Creepy-crawly,” “Jittery,” “Pulling,” “Shock-like pain,” “Burning,” “Throbbing,” “Tightness,” “Itching,” etc. (7), with pain being the primary symptom among over 50% of RLS patients (8). The sensations are usually experienced deep in the leg along with the corresponding sense of movements (7). The symptoms of RLS often degrade normal daily function but are improved with movement (5). RLS is prevalent in 5–10% of the general population (6), although studies have reported rates as high as 11% (9) and 15% (10).

Patients with RLS tend to seek medical care due to impaired sleep (11). The current diagnostic protocol relies primarily on subjective questionnaires because the physical and electrophysiological examinations typically appear normal in RLS. Therefore, the condition is often misdiagnosed. In fact, a study utilizing nerve conduction velocity testing found that there was no difference between those with and without RLS among those with diabetic peripheral neuropathy (12). Other conditions such as iron deficiency and comorbid neurological conditions, including peripheral neuropathy and radiculopathy, are commonly associated with RLS.

The two commonly described forms of RLS are primary and secondary. Approximately two-thirds of RLS patients present with the primary form (13), which generally presents with an onset prior to 45 years of age and has a strong genetic/familial component (14, 15). The familial component is strong among those with idiopathic RLS but not among those with RLS and peripheral neuropathy (15) or secondary RLS.

When diagnosed later in life, RLS is often secondary to other neuropathic disorders, associated with a comorbid condition (16), and more symptomatic than those with primary RLS (17). Other neurological conditions often found in conjunction with RLS include Parkinson’s disease (18), multiple sclerosis (19), myelopathies (20), essential tremor (21), and peripheral neuropathy (16, 22, 23).

Due to accompanying comorbidities, RLS is often misdiagnosed or treated merely for symptoms (24). Furthermore, successful treatment is complicated by a poor understanding of the etiology and mechanisms of RLS. Clinical evaluation should attempt to distinguish primary vs. secondary RLS.

Prescription treatment of the strongly familial primary RLS includes dopamine agonists [ropinirole, pramipexole (25), rotigotine (26), cabergoline, levodopa]; alpha 2 delta ligands (gabapentin enacarbil (27), pregabalin (28)), opioid agonists (oxycodone/naloxone), and iron. Additional physical treatment methods include near-infrared spectroscopy, pneumatic compression, transcranial direct current or magnetic stimulation, and vibrating pads.

Secondary RLS can be treated with prescription of ropinirole, levodopa, vitamin C and E supplementation, or exercise. However, this treatment strategy is only efficacious for those with end-stage renal disease on hemodialysis (24). End-stage renal disease can reduce erythropoietin levels. The resulting anemia can cause nerve damage and possibly induce secondary RLS.

Reports have shown an increased prevalence of RLS among those with various forms of peripheral neuropathy, ranging from 5.2 (58) to 54% (59). A few early studies found no difference in prevalence between neuropathy patients and controls (60–62). However, this may be due to less strict identification criteria prior to the validation and wide acceptance of the IRLS Study Group rating scale (63) or differences in the specificity or heterogeneity of the neuropathy [i.e., general peripheral neuropathy vs. diabetic sensory polyneuropathy (DSPN)] (64). Peripheral neuropathy may be considered an umbrella term that covers many forms of paresthesias of different etiology. This includes DSPN, demyelinating neuropathy, inflammatory demyelinating polyradiculoneuropathy, Fabry’s disease with neuropathy, hereditary neuropathy, small fiber painful neuropathy, and neuropathy due to Charcot-Marie-Tooth disease, all of which have been reported in conjunction with RLS (12, 17, 65–68). DSPN, however, is most frequently associated with RLS (17).

One study reported a greater prevalence of RLS among diabetic patients (18–29%) compared with controls (6–7%) (13, 16). The presence of polyneuropathy was the only variable associating RLS with diabetes and not the metabolic attributes of diabetes (16). However, a follow-up case-controlled study found the prevalence of RLS in diabetics to be independent of peripheral neuropathy (69). RLS in diabetic neuropathy appears to be independent of iron levels (16, 62), sex, and duration of diabetes (16). Diabetic neuropathy is often associated with damage to the small nerve fibers, which led to speculation of a small fiber mechanism associated with DSPN and RLS (70). This small fiber involvement distinguishes “peripheral” RLS from “central” RLS (71), which is thought to have a dopaminergic mechanism (72). However, others hypothesized that diabetes may inhibit dopaminergic control at the spinal level via excitatory nociceptive inputs associated with peripheral neuropathy (16).

Reports also showed that 27% of RLS patients had underlying peripheral neuropathy, and of those 34% had primary RLS (59). Those with peripheral neuropathy and RLS have higher IRLS scores, greater symptom severity, and more disturbed sleep compared with those with primary RLS (17). Studies have found that 92% of those with idiopathic RLS without neuropathy had a family history of RLS as opposed to only 13% of RLS patients with neuropathy; therefore, there may be a difference in the etiologic nature of primary and secondary RLS (15).

Emerging research in surgical decompression has found that the symptoms associated with polyneuropathy may be reversible by decompressing focal nerve entrapment sites in the lower extremity. Entrapment neuropathies constitute around 30% of all diabetic neuropathies (73). Peripheral nerve entrapment is associated with nerve enlargement, whereby those with diabetic neuropathy have tibial nerve cross-sectional areas twice as large as those of diabetics without neuropathy and non-diabetic controls (74). The enlargement of the posterior tibial nerve has been shown to significantly improve following surgical decompression (75). Surgical decompression in the lower extremity is similar to that of the surgical decompression of the median nerve at the carpal tunnel, whereby the neural entrapment is relieved by release of the fascia constituting the nerve tunnel. Reports have described a disorder analogous to RLS in the upper extremity, whereby carpal tunnel resection either improved or resolved the restlessness of the upper extremity (76). This implies that there may be a peripheral nerve tunnel entrapment component to RLS symptoms.

The subjects were included through a retrospective analysis of patients who underwent decompression of the common and superficial fibular nerves for either diabetic (n = 14) or non-diabetic (n = 28) peripheral neuropathy. All surgical procedures were performed by Dr. James Anderson, DPM, at Anderson Podiatry Center for Nerve Pain. Eleven men and 31 women were included. All of the subjects were diagnosed with RLS by a physician prior to surgery through patient history and clinical presentation. Patients were also queried on impaired sleep quality, daily fatigue, fall risk, and location of perceived symptoms during the initial clinical assessment. Muscle strength (manual muscle test), physician-observed gait quality, and proprioception (Romberg’s test) were assessed during the physical examination. The self-reported average age of the patients was 63 ± 11 years at the time of RLS diagnosis. Surgical eligibility for peripheral neuropathy was based on medical history and physical examination, including visual analog scales (VAS) >6 and positive Tinel’s sign in the affected nerve tunnels (79). See Anderson (80) for a detailed description of the clinical examination and surgical decompression.

Benefits, risks, and alternatives were presented, and informed consent was obtained from all subjects prior to surgery. Forty-two subjects underwent nerve decompression surgery between October 23, 2013, and December 9, 2015. The protocol and retrospective analysis were approved by the University of Colorado Health Institutional Review Board.

Subjects scored their symptoms for the surgical leg on a scale of zero (no symptoms) to 10 (severe symptoms) without the use of medication. Scores were obtained during normal clinical visits both before surgery and 15 ± 5.6 weeks post-surgery for the following symptoms: pain, burning, numbness, tingling, weakness, balance, tightness, aching, pulling, cramping, twitchy/jumpy, uneasy, creepy/crawly, and throbbing.

The surgical decompression was performed in accordance with previously described methods for both the common (80) and superficial fibular nerves. Some subjects also underwent decompression of the deep fibular (57%), tibial nerve at the tarsal tunnel (14%), or tibial nerve at the soleal sling (57%) entrapment sites, depending on their clinical presentation.

Repeated measures analysis of variance was used to assess the differences between the pre- and postsurgical VAS ratings for each symptom. Total pre-surgery and total post-surgery VAS scores were calculated by summing the individual symptom categories. The change in total VAS was calculated (post-surgery minus pre-surgery). Correlations between pre-surgery VAS and the change in VAS were also calculated. To account for the multiple pre–post comparisons of symptoms, a more stringent P < 0.01 value was used to determine statistical significance. An additional analysis was conducted to account for the multiple comparisons using the Holm–Bonferroni Method (81).

All of the individual symptoms exhibited significant improvement following decompression (Figure 1), with the exception of “Pulling” (P = 0.14). The changes in the Weakness (P = 0.019) VAS score did not reach the P < 0.01 level of significance. The Holm–Bonferroni analysis produced P-value– ≤ 0.01 for all of the VAS scores, except Pulling (P = 0.05). The change in VAS score was negatively correlated with the pre-surgery VAS score for both the total summed VAS (r = −0.58, P < 0.001, Figure 2) and all of the individual symptom VAS scores (all P < 0.01).

There were no differences between men and women or between diabetics and non-diabetics for the change in VAS score after surgery (P > 0.05).

Due to the retrospective nature of this study, it did not include an assessment of other key attributes of RLS, such as iron deficiency. The relationship between RLS and iron deficiency remains unclear as some studies have found RLS to be independent of iron deficiency among diabetic patients (62), whereas others have found no statistical difference between hematocrit and serum iron and ferritin levels among those with primary RLS compared with neuropathy/secondary RLS (17).

The main finding from this retrospective analysis is that surgical decompression of the common and superficial fibular nerves improved patient ratings of symptoms commonly associated with RLS. This preliminary finding suggests the need for rigorous, controlled, prospective research on the potential for nerve decompression surgery as a viable treatment option for RLS patients.

The total change in VAS was negatively correlated with the total presurgical VAS. Therefore, those who presented with the most severe symptoms experienced the greatest improvement following nerve decompression. This report is limited in that the data were collected retrospectively without a detailed collection of RLS-specific patient characteristics, including detailed differentiation between primary and secondary RLS. Nevertheless, these patients underwent decompression surgery for the treatment of peripheral neuropathy symptoms and also exhibited improvement in RLS-related symptoms, which is a finding that warrants additional investigation.