Restless Legs Syndrome in the Darkness of Dementia

Restless Legs Syndrome (RLS), also known as Willis-Ekbom Disease, is a common sensorimotor sleep-related movement disorder characterized by an “urge to move the legs” and an “unpleasant sensation.”¹ Due to the circadian variation of symptoms, these begin or worsen during periods of rest, mainly during the evening or night. RLS is associated with severe clinical consequences including sleep-onset or sleep-maintenance insomnia, depression and anxiety, possible increased vascular risk, and other conditions.² As a result, other important areas of functioning including occupational, educational, and other daily activities could be negatively influenced. Prevalence of RLS ranges from 2.4% to 10.8% in Caucasian populations, while the lower prevalence rates have been reported in East Asia and South America.³ Idiopathic RLS, clinically similar to secondary RLS, is usually familial and has an early onset (before the age of 30 years). Secondary RLS is associated with various conditions, including iron deficiency, kidney failure, and pregnancy. Currently there are two first-line treatment choices, namely dopamine agonists and α2δ ligands.³ Although dopaminergic dysfunction, abnormal iron metabolism, central opiate system, and genetics have been implicated, RLS pathogenesis largely remains unknown.⁴ Diagnosis of idiopathic or secondary RLS is based on five essential clinical criteria that were recently updated.¹

In this issue of SLEEP, Richards and colleagues try to face a very challenging problem: identifying a condition that is diagnosed on clinical grounds in persons with dementia who cannot self-report.⁵ They try to bypass this inherent difficulty by creating a behavioral observation test (Behavioral Indicators Test - Restless Legs, BIT-RL) in addition to clinical measures and physiological recording (Periodic Activity Monitor-Restless Legs, PAM-RL). In the absence of such instruments this is a very important first step towards developing this field. Their efforts are even more significant given that sleep disturbances are extremely common in dementias, and that persons with dementia have a number of behaviors possibly indicative of underlying RLS.^6,7 At the same time, it is quite clear that the diagnosis of RLS in patients with dementia still has many limitations and obstacles, the main one being validation of the instruments in the target population of demented subjects. Both dementia and RLS diagnoses remain clinical. In the absence of valid biomarkers, there is ample room for diagnostic error in both conditions, making the elucidation of true relations between RLS and specific dementias quite challenging. Along this line, it may be helpful if future attempts to validate assessment approaches in specific types of dementia types (i.e., Alzheimer disease). It would be an additional step forward if diagnosis of RLS in dementia is biomarker based. This may provide more information regarding association of specific RLS-biological changes (e.g., ferritin level) with specific biological pathways (e.g., β-amyloid, tau).

Foreseen challenges in this future validation process include the following: (1) There is an extremely high, almost universal use of medications that have been associated with RLS in demented subjects (e.g., neuroleptics, SSRIs). (2) Many of the indicators included in BIT-RL⁵ are very common in dementia, particularly in the setting of neuropsychiatric symptoms such as aggression, agitation, irritability⁸ and even depression.⁹ (3) As RLS symptoms usually occur or worsen at night, many of the dementia-associated neuropsychiatric symptoms also exacerbate late during the day in period of sundowning.⁴ Some of the subjective sleep complaints, the strongest predictors of RLS in this report,⁵ may not be accurately reportable by dementia subjects (e.g., difficulty falling asleep, daytime fatigue, leg discomfort) and may not be readily obtainable by informants (e.g., discomfort in legs). Additionally, some of them (difficulty falling asleep, daytime fatigue) maybe confounded by coexisting apathy and depression. As a result, more weight has to be placed on the developed observational instruments.

More research on qualifications and characteristics of trained raters who can accurately score the BIT-RL would be helpful. For example, it would be quite informative to explore whether dementia informants and caregivers (not uncommonly spouses, thus elderly subjects) could provide valid and reliable BIT-RL ratings, or whether short observations by expert clinicians would be necessary.

Apart from the necessity of an accurate diagnosis of RLS in demented patients, evaluation of RLS severity is also crucial since not all RLS patients need to be treated. In clinical practice, the severity of RLS is often evaluated using rating scales such as the International Restless Legs Rating Scale, and treatment is prescribed mainly in patients with moderate and severe RLS. This should be another important focus of future research in order to avoid overtreatment of RLS in demented patients.

It is also essential to determine whether a proposed instrument for diagnosing RLS will also be useful for measuring treatment response, and for identifying augmentation (i.e., worsening of RLS symptoms compared to those before treatment initiation), which is a common treatment-related side effect in patients treated with dopaminergic agents.¹⁰

In summary, the creation and validation of BIT-RL⁵ should be considered as the first important step toward better understanding and unraveling the complexities of diagnosis and treatment of RLS in demented patients.

CITATION

Hadjigeorgiou GM, Scarmeas N. Restless legs syndrome in the darkness of dementia. SLEEP 2015;38(3):333–334.

DISCLOSURE STATEMENT

The authors have indicated no financial conflicts of interest.