Discussion

The primary intent of this report was to conduct a comparative effectiveness review on treatments for restless legs syndrome. However, we identified only two randomized controlled trials that directly compared treatment options. Included studies did not permit reliable indirect comparisons from which to draw robust conclusions about comparative benefits and harms. Results from small, placebo-controlled randomized trials of generally short duration demonstrated that dopamine agonists (ropinirole, pramipexole, and rotigotine) and anticonvulsant alpha-2-delta ligands (gabapentin enacarbil, gabapentin, and pregabalin) increase the percentage individuals responding to treatment (as defined by a 50 percent reduction in the International Restless Legs Syndrome (IRLS) Study Group symptom scale score or reporting “improved or much improved” on the clinician-assessed global impressions scale score (CGI) or patient-assessed global impressions scales score (PGI), reduce restless legs syndrome (RLS) symptoms, and improve disease-specific quality of life and patient-reported sleep outcomes. However, adverse effects of pharmacologic therapies and long-term treatment withdrawals due to adverse effects or lack of efficacy are common.

Evidence is lacking about the long-term effectiveness in, and applicability to, adults with less severe or less frequent RLS symptoms, children, or individuals with secondary RLS including those with iron deficiency, end-stage renal disease, or pregnant women or those intending to become pregnant. Studies of pharmacologic therapies consisted mainly of dopaminergic agents; a few studies assessed alpha-2-delta ligands. All studies administered therapies daily rather than “as needed.” Although the effectiveness, harms, and adherence to “as needed” therapy are unknown, current recommendations note this as an option.6 Few nonpharmacologic therapies were assessed, and no individual nonpharmacologic treatment was studied in more than a single trial. Randomized controlled trials (RCTs) were short in duration and enrolled highly selected populations with symptoms that were very severe to high-moderate, frequent, and long-standing. Additional meta-analyses are supportive of our findings.8486

Exclusion criteria were many, and subjects were typically recruited from RLS clinics rather than primary care or mental health settings; both settings are frequent sites for initial detection and management of individuals with RLS. Enrollees had greater disease severity, frequency, and duration than was reported by the estimated 1.5 percent of individuals described as “RLS sufferers” based on a telephone survey of adults who agreed to be interviewed about RLS. No RCTs assessed patients with mild or moderate disease, and few lasted longer than 6 months. None enrolled individuals under age 18, and the vast majority of individuals were white.

We included studies that reported validated RLS symptom scale measures assessing overall disease severity, impact, quality of life, patient- and physician-reported global assessment, and sleep quality. However, thresholds establishing a clinically important effect size are unknown. Although symptom scales are widely used in research studies, their use in clinical settings is less clear and likely limited. Furthermore, despite the fact that RCT study subjects met consensus definitions of RLS, these criteria may not be routinely used in clinical settings to diagnose, assess severity, or initiate therapy. Thus, we do not know the applicability of results from these RCTs to individuals seen, diagnosed and treated in primary care or mental health settings. Outcomes were not stratified by patient and RLS characteristics, and we could not determine whether findings vary by these factors. Other scale scores are often reported. We focused on outcomes that are most widely used, appear to have the greatest face validity and have clinically meaningful impact especially relevant to patients diagnosed and treated in the United States.

Only two RCTs directly compared pharmacologic options; specifically cabergoline to levodopa and pramipexole to dual-release levodopa/benserazide. We found no clear evidence of a dose effect for the outcomes of IRLS responders and mean change in IRLS scale scores for either dopamine agonists (k=3) or the alpha-2-delta ligands (k=2). Because studies reported a large placebo response, we urge caution in using information from uncontrolled studies as the basis for recommending increasing drug doses or altering administration timing if symptom response is inadequate. Similarly, we urge caution in attributing benefits that might be observed in clinical settings to dose adjustment. One study comparing pramipexole versus pregabalin has recently been completed and is expected to be published shortly.

Few studies assessed individuals with secondary RLS. No studies enrolled pregnant women. Only two studies assessed the effect of iron therapy on RLS symptoms in adults with iron deficiency. These studies were small, short, and had methodological flaws; however, they suggested that iron therapy may improve symptoms in these individuals. A single study that did not meet our eligibility criteria because it did not use validated RLS symptom scale scores found no benefit with oral iron therapy in adults with RLS and normal iron stores.87 Another small short-term RCT assessed intravenous iron versus placebo in patients on hemodialysis with normal iron stores. This study found no benefit. We identified one other study in adults with RLS believed secondary to end-stage renal disease. This study compared gabapentin to placebo, did not report validated RLS symptom scale scores, and showed no benefit with the drug.

For individuals unable to initiate or tolerate dopaminergic agents, or for whom these drugs have failed, recommended pharmacologic treatments include off-label opioids (morphine, oxycodone and methadone), sedative hypnotics, and tramadol. None of these are FDA approved for treatment of RLS and all have the potential for long-term abuse especially given the subjective nature of RLS symptoms and the large placebo response seen in other pharmacologic studies. We found no eligible studies evaluating these agents. A single crossover study of 11 patients assessed oxycodone versus placebo and reported improvement in leg sensation, motor restlessness, and alertness.

Randomized controlled studies should be initiated to evaluate the benefits of these therapies not approved for treatment of RLS in individuals who are refractive to standard pharmacologic treatment.

We found no data from RCTs on the comparative benefits or harms of dopamine agonists and anticonvulsant alpha-2-delta ligands. Only two small studies of iron therapy addressed secondary RLS due to iron deficiency, providing low strength of evidence that iron replacement therapy may improve symptoms. Assessment of nonpharmacologic interventions was limited to four trials. These provided low-strength evidence for a benefit with compression stockings, near infrared light, and exercise, but not for valerian.

No studies assessed the effect of patient characteristics on treatment benefits and harms. We found no evidence on effectiveness of these interventions in children, older adults with multiple morbidities, pregnant or recently postpartum women, or individuals with end-stage renal disease. All pharmaceutical trials were industry sponsored. No studies meeting our inclusion criteria assessed opioids, sedative hypnotics, or tramadol, all of which are recommended in treatment algorithms6 and presumably used in clinical practice.

Trials reported a large placebo effect, thus future studies require adequate blinding. Moreover, clinicians and patients should be aware of such a large placebo response. Applicability is limited to nonpregnant adults who have high-moderate to very severe RLS and no major comorbidities. Long-term studies reporting withdrawals due to loss of efficacy or side effects suggest that for many RLS patients, the benefits of pharmacologic treatment are not sustained over time, and that these treatments result in adverse effects and are often discontinued. Augmentation, a drug-induced exacerbation of the disease, can occur with dopaminergic drugs.

Evaluating RLS treatments requires determining the change in scale scores that constitutes a minimum clinically important difference. These thresholds have not been established for the IRLS scale score and other scales commonly reported in RLS research. Further, high-quality research is needed to determine whether treatment benefits observed in short-term studies are maintained, and whether the therapies are tolerated long term. The target populations for these drugs are patients with moderate to severe RLS, who may require daily treatment for decades. Even nonpharmacologic interventions and other treatments for those with milder symptoms are often long term. Yet, evidence is limited to short-term efficacy trials or observational studies among highly selected individuals.

Given such limited evidence, patients and providers face uncertainty regarding the benefits and risks of RLS treatments for individuals whose symptoms are less severe, less frequent, of shorter duration, or diagnosed based on criteria that differ from RLS consensus definitions. Results from short-term efficacy trials in highly selected population of RLS patients should be carefully interpreted for their applicability to the more heterogeneous population of RLS patients in primary care settings. Applicability concerns are even more salient in light of direct-to-consumer marketing that has raised awareness of potential RLS symptoms. The populations in clinical trials had RLS of high-moderate to severe intensity for many years, and many of these patients had received previous unsuccessful drug treatment for RLS. In contrast, individuals presenting to primary care with RLS like-symptoms may have milder symptoms or other conditions whose symptoms mimic RLS (e.g., periodic leg movement disorders, nocturnal leg cramps, vascular or neurogenic claudication). They may also be younger, older, or have more comorbidities than subjects included in available RCTs.

In conclusion, randomized controlled trial evidence for RLS treatments is mostly limited to short-term, placebo-controlled studies of dopamine agonists and alpha-2-delta ligands conducted in a highly selected population of adults with moderate to very severe primary RLS of long-duration. Compared to placebo, dopamine agonists and alpha-2-delta ligands increase the percentage of individuals “responding,” reduce RLS symptom scores, and improve patient-reported sleep outcomes, disease-specific quality of life, and overall RLS impact. Both short- and long-term adverse effects and treatment withdrawals due to adverse effects or lack of efficacy for dopamine agonists and alpha-2-delta ligands are common. We found no high quality data on comparative effectiveness and harms of commonly used treatments, little data on nonpharmacologic interventions or the effect of patient or RLS characteristics on outcomes. Applicability is unknown for adults with less frequent or less severe RLS symptoms, children, or those with secondary RLS.

Future Research Recommendations

Table 15 summarizes our main recommendations for future research based on the gaps identified in this review.

Table 15. Future research recommendations.

Table 15

Future research recommendations.