Diagnosis of Restless Legs Syndrome

Published on 12/04/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 2101 times

Chapter 15 Diagnosis of Restless Legs Syndrome

Restless legs syndrome (RLS), like most syndromes, represents multiple biological pathologies that share a common clinical presentation. As such, the diagnosis of RLS relies almost entirely on the subjective report of symptoms matching the defining features of RLS. Nonetheless, the nonspecific but sensitive objective motor signs of RLS—periodic limb movements while lying resting awake (PLMW) and while asleep (PLMS)—provide important support for the diagnosis. Given the critical nature of the clinical presentation, after a brief review of the historical development of the diagnosis, this chapter explores the clinical aspects of the defining features. The four clinical features that define RLS appear to be simple enough, but clinicians commonly misunderstand them and fail to appreciate some of their specific expressions. Accurate diagnosis starts with a full understanding of these four defining features as they manifest in RLS. The diagnosis can be further aided by three supportive features of RLS and by recognition of differing RLS phenotypes. This chapter ends with the presentation of tools that have been developed to support making the RLS diagnosis.

History

What appears to be the earliest description of RLS in the medical literature by Willis ^1,² in the 17th century emphasizes both the abnormally excessive movements and their occurrence during usual sleep times. References to RLS after Willis appear to largely assign RLS to a psychiatric or psychological disorder (e.g., related to anxiety³ or, more specifically, financial worries⁴) until the middle of the 20th century, when Ekbom⁵ described the presentation of the disorder in a large series of cases. His work both provides the name currently used for the disorder and establishes RLS as a neurologic disorder. Ekbom emphasized the sensory aspects (paresthesias in the legs) of the disorder⁶ more than the motor features (contractions of the legs) described by Willis. Even after this excellent work, RLS remained largely ignored until the latter part of the 20th century, when a small group of clinicians treating RLS patients formed the International Restless Legs Syndrome Study Group (IRLSSG) and developed the first clinical consensus on the diagnosis of RLS.⁷ These diagnostic criteria restored the emphasis on the movement aspects of RLS and in particular noted an urge to move the legs often but not always associated with the paresthesias. There remained some confusion in the criteria established by this group, particularly in relation to the concept of “restlessness.” These issues were resolved in a National Institutes of Health workshop where the current diagnostic standards were developed.⁸ Thus, as shown in Box 15-1, the RLS diagnosis has evolved from observed movements to the recognition of the akathisia focused in the legs and modulated by diagnostically significant factors.

BOX 15-1 Evolution of Restless Legs Syndrome (RLS) Diagnosis

Willis (1685)	Movement and sleep disruption described
Ekbom (1945)	Sensory disturbance emphasized
American Academy of Sleep Medicine (AASM) (1990)	Sensory and periodic leg movement disturbance of sleep, nocturnal worsening emphasis
International Restless Legs Syndrome Study Group (IRLSSG) (1995)	Restores emphasis on movement as “restlessness”
Restless Legs Syndrome Foundation/IRRLSG/National Institutes of Health Workshop (2002)	Final formulation of diagnostic standards
	Emphasis on “urge to move,” removes motor restlessness and separates effects of rest and activity; clears up some concepts related to supportive features

Essential Diagnostic Criteria

All four of the diagnostic criteria given in Box 15-2 must be met to make the diagnosis of RLS, and failure to meet any one of these excludes the RLS diagnosis. These four criteria appear deceptively simple, but further examination reveals several subtle aspects embedded in these diagnostic features.

Criterion 1: Urge to move the legs usually but not always accompanied by an unpleasant or uncomfortable sensation in the legs. This first essential diagnostic criterion defines the basic symptoms of RLS, which must include a strong urge to move the legs or what is best described as an akathisia focused in the legs. This focal akathisia may occur in either one or both legs and may be focused in only a part of the leg or foot or may extend to include other parts of the body, but it must involve the leg. This is a conscious sensation of an urge to move the leg and not an unconscious habit such as foot tapping when sitting. Movements made without awareness occur with RLS as rather abrupt involuntary physiologic flexors of the legs similar to the periodic leg movements of sleep, but these do not involve a sensation of the urge to move and do not suffice for the diagnosis of RLS. The patient reporting that others note they move a lot or even notice themselves to be moving a lot are not reporting a symptom that meets this criterion. This criterion requires the patient to be fully aware of the urge to move and describes that sensation as localized to or arising from the legs.

The focal akathisia also occurs over a significant duration of time lasting at least a few minutes unless relieved by actions reducing the akathisia, usually moving the legs or other affected body part. The momentary discomfort due to an uncomfortable posture that can be relieved by postural change should not be confused with the need to move of RLS.

The importance of ensuring the presence of this basic aspect of RLS cannot be overemphasized. For example, conditions have been reported that appear to have all the features of RLS without the focal akathisia but with the involuntary leg movements of RLS. These are described as quiescegenic nocturnal dyskinesias and do not show the same clinical pattern as RLS; in particular, they are not universally responsive to dopaminergic treatment.

This diagnostic criterion also includes abnormal and unpleasant sensations in the legs that usually, but not always, accompany the akathisia, but when they occur, they are associated with or even seen as causing the akathisia. These sensations, although decidedly unpleasant, are generally not seen as painful, although in some case series of clinical patients, 30% to as many as 56% report that symptoms are painful.⁹ The type of pain reported, however, is not the severe hurting pain of a leg cramp, for example, but more like an ache such as a toothache. Because these RLS sensations arise without any detectable abnormality in the leg, they have no common intersubjective reference and patients find it very hard to describe these symptoms. They usually say the sensations are like some imagined event, such as worms crawling in the veins. Box 15-3 lists some of the common subjective patient reports of these sensations. Although psychological methods to better define the characteristics of these abnormal sensations exist, they have not yet been used. Nonetheless, the sensations appear to have two basic dimensions. First, they are more dynamic than static; that is, they often involve a sense of something moving through the leg or in the leg and they are not like a point source of pain. Second, they are deep in the leg, often described as in the bone and not on the skin or surface of the leg. Thus, one description is an “itch in the bone.”

BOX 15-2 Essential Restless Legs Syndrome (RLS) Diagnostic Criteria

All four of the following criteria must be met to make the diagnosis of RLS:

1. An urge to move the legs occurs, usually accompanied or caused by uncomfortable and unpleasant sensations in the legs.

2. The urge to move or unpleasant sensations begin or worsen during periods of rest or inactivity, such as lying down or sitting.

3. The urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues.

4. The urge to move or unpleasant sensations are worse in the evening or night than during the day or only occur in the evening or night.

The next three criteria define features that modulate the expression of the RLS symptoms defined by the first criterion. The second diagnostic criterion defines the features that engender RLS, whereas the third criterion defines those that reduce the symptoms and the fourth defines the strong circadian pattern of symptom expression.

Criterion 2: Rest engenders RLS symptoms. Rest here involves two features: decreased movement occurring when sitting or lying down and also decreased mental activity. These provide the conditions that engender RLS symptoms. The suggested immobilization test (SIT) relies on this aspect of RLS (see Chapter 18). In this test, the subject is to stay awake and not move while sitting upright in bed with the legs stretched out in an environment devoid of any changing stimulation. This creates the rest condition provoking the RLS symptoms. The symptoms are expressed by both the occurrence of the urge to move and the periodic limb movements (PLM) representing either involuntary movements or a response to this urge that cannot be suppressed under these conditions. Studies have shown that as the duration of the rest gets longer, the RLS symptoms are more likely to both occur and become more severe. This validates the diagnostic relation between rest and RLS.¹⁰^–¹² Thus, rest acts as a stimulus producing the symptoms that increase in strength with the degree and duration of the rest. It is important to recognize that it is any rest situation that produces the RLS symptoms and not a particular resting body position. Thus, symptoms that occur only with a particular body position such as sitting with the legs crossed are not likely to represent those produced by RLS. Any rest that lasts long enough should engender the symptoms. Because lying down represents a more restful position than sitting, some subjects may report symptoms only when lying down and may not observe them when sitting.

Some subjects with milder RLS may require a longer period of rest before the symptoms are expressed, and if they fall asleep quickly, they may therefore experience their symptoms mostly when sitting for a while and only rarely when lying down. The RLS diagnosis should be questioned whenever the patient reports symptoms only when sitting and hardly ever when lying down unless the symptoms are mild and occur only after a relatively long duration of sitting.

It is generally reported that increased confinement and decreasing general movement enhances the effects of rest. Thus, symptoms are commonly reported to occur when sitting in airplanes, cars, or theaters, even in patients who rarely have the symptoms at other times.

Criterion 3: Leg movements relieve RLS symptoms. The degree of relief appears related to the degree of leg movements. Simply moving the legs while lying down may help some, but more relief is provided by standing and even more by walking. Two features of the relief from movement are often missed. First, the relief should at least be partly felt almost immediately or very soon after the movement is started. The relief occurs because of the movement and not as a result of exercise or intensity of activity. Some patients may have trouble identifying how long it takes for movement to relieve the symptoms, but most, when carefully questioned, recognize the rapid, at least partial relief occurring immediately with or shortly after the onset of the movement. Second, the relief occurs while moving and should persist for as long as the movement continues. The immediacy and persistence of the relief with movement aid in the diagnosis of RLS.

Some patients may confuse relief persisting after movement stops with that during the movement. When asked if symptoms are relieved by movement, they may say something like, “No, the symptoms come back when I stop moving.” But the report of symptoms returning when movement ceases indicates they are less during movement.

It deserves note that movement is only one method to relieve RLS symptoms. Subjects report that any intense conversation or argument, intense mental involvement such as in a computer game, eating, hard rubbing of the legs, and very hot baths also reduce symptoms. As shown in Table 15-1, the activities reported to reduce RLS symptoms are identical to those producing alertness. Conversely, situations decreasing alertness engender the RLS symptoms. In this sense, RLS can be seen as a disorder of the quiet resting state with decreased alertness producing symptoms and increased alertness reducing symptoms. Movement provides the most consistent method for increasing alertness, and thus it has been recognized as critical for the diagnosis; it is not simply movement, but rather the motor activation and mental alertness that also occur with the movement that reduce the symptoms. Standing on one leg may be as good or even better than actually moving the leg because it produces a significant degree of alertness; at least, it is hard to sleep while standing on one leg for a period of time, although it may be possible to sleep even while walking.

Criterion 4: Nocturnal exacerbation of RLS symptoms. SITs repeated during the daytime document the increasing severity of symptoms ordered from morning to afternoon to evening.¹² Thus, the worsening of symptoms in the evening and night results not from decreased activity in the evening but rather from some aspect of the circadian biology of man. Reduced symptoms in the morning even occur when the patient has not been asleep.^13,¹⁴ The circadian pattern provides a strongly protected period in the morning from about 8 A.M. to 10 A.M. with few RLS symptoms. This asymmetry in the circadian modulation affects temporal expression of symptoms such that as RLS becomes worse, symptoms expand into the afternoon, while the morning times remain relatively free from symptoms. When enquiring about the circadian pattern of symptoms, care should be taken to ask about symptom severity when the patient is resting at various times of the day, avoiding the confounding effect of more rest in the evening. The clinician should also seek to ascertain that the patient can sit or lie down in the morning with relatively few RLS symptoms unless the RLS is very severe with intense symptoms in the evening and night.

Patients with circadian rhythm disturbance or who are shift workers may have an underlying circadian phase differing from the normal pattern because RLS symptoms generally occur in relation to the underlying circadian phase. Some patients with mild RLS may experience symptoms only when sitting for a while and may report their symptoms are worse in the evening than at night, because the duration of rest while awake at night is too short to provoke the symptoms and the symptoms are too mild to wake the patient. Conversely, patients with very severe RLS symptoms may have lost all circadian variation with severe symptoms occurring around the clock, but they should be able to describe a time in the past when the symptoms were less severe and occurred more in the evening or night than in the morning.

TABLE 15-1 Activities to Reduce Restless Legs Syndrome (RLS) Symptoms Compared With Those to Stay Awake

Reduce RLS Symptoms	Stay Awake
1. Movement	1. Movement
Sitting up helps very little	Sitting up helps very little
Standing helps some	Standing helps some
Walking, moving help a lot	Walking, moving help a lot
2. Intense conversation—argument	2. Intense conversation—argument
3. Eating food	3. Eating food
4. Hard rubbing of the legs (or arms)	4. Hard rubbing of the body
5. Very hot or cold baths	5. Bracing (hot or cold) shower or bath

Features Supporting the Diagnosis

In addition to the essential diagnostic criteria, there are three features of RLS presented in Box 15-4 that provide support for the diagnosis. These are particularly helpful for establishing a diagnosis when there is some uncertainty about the essential diagnostic criteria. They also inform about the disorder, and one provides a method for objective evaluation of change in symptom severity.

Levodopa or dopamine agonist treatment decreases RLS symptoms in almost all cases. This treatment benefit may not persist, but the patient should be able to describe at least an initial, if only temporary, benefit from the dopaminergic treatment. This may not occur if the dose of the medication is not raised to the usual therapeutic range for RLS. In general, if a patient reports no symptom improvement with dopaminergic treatment, the diagnosis should be re-examined.¹⁵ Clinical experience indicates this rarely happens.

PLMS and PLMW occur in at least 80% to 85% of RLS patients.¹⁶ PLM represent the motor sign of RLS. PLMS also occur in several other neurologic conditions, mostly those involving the dopaminergic system,¹⁷ and occur with the use of most antidepressants, particularly venlafaxin and the selective serotonin reuptake inhibitors.¹⁸ They appear to occur more with advancing age, particularly in those over 55, although this increase occurs more in families selected because one member has RLS than in families selected because one member does not have RLS or PLM.¹⁹ Thus, these PLM may be a forme fruste of RLS, or they may represent a genetic factor that commonly occurs with the RLS. Moreover, the PLMW during a night’s sleep are rare in older adults. In one study examining the diagnostic usefulness of PLMS and PLMW from the standard polysomnogram (PSG) and from the evening SIT, the PLMW provided the best diagnostic usefulness, with a sensitivity of 87% and specificity of 80%.²⁰ If these values are confirmed in larger studies, the PLMW from the night PSG can be seen as providing a useful diagnostic test. Certainly the absence of PLM (either PLMS or PLMW) should lead to careful reconsideration of the clinical diagnosis (see Chapter 18).

Two separate studies using two different clinical scales reported a significant correlation between PLMS and subjective evaluation of clinical severity of RLS.^21,²² PLMS also correlated with sleep efficiency for RLS patients. Thus, this motor sign of RLS provides both a useful diagnostic support and a valuable objective assessment of the clinical severity. The latter is particularly useful for treatment evaluations.

The PLMS can also be determined from leg activity monitors with body position sensors built into them to separate lying down from other body positions. These monitors provided by IM Systems (Baltimore, MD) permit multiple nights’ assessment of PLMS. This reduces the error introduced into the assessment by the high night-to-night variability in PLMS for RLS patients,²³ and adding extra nights has been shown to increase the diagnostic precision for PLMS.¹⁰ (See Chapter 19 for further consideration of activity monitoring in RLS.) It may be that further exploration of the PLM features eventually leads to even better diagnostic usefulness for this measure.

Family history of RLS commonly occurs particularly in patients whose RLS symptoms started occurring early in life before age 20⁹ and also before age 45.^24,²⁵ The absence of the first two supportive features indicated RLS diagnosis was not likely; in contrast, the presence of this third supportive feature increases the likelihood of the diagnosis and its absence is not particularly helpful. In one case series of primary RLS, almost 50% reported not having a first-degree relative with RLS.²⁶ The lack of any relatives with RLS has no significance for RLS diagnosis, particularly for the patient with RLS symptoms starting after age 45. This may not be the case for pediatric RLS; PLM actually becomes a significant factor contributing to the diagnosis of RLS in children.⁸

BOX 15-4 Features Supporting the Diagnosis of Restless Legs Syndrome (RLS)

Response to dopaminergic therapy

Periodic leg movements (during wake or sleep)

Family history of RLS

Phenotypes of Restless Legs Syndrome

RLS as a syndrome is likely to have multiple causes, and there may be subtle differences in the clinical presentation reflecting the differing pathologies. There are also some clearly defined conditions causing RLS. Moreover, the possibility of partial expression of the syndrome deserves some consideration in situations where there is reason to expect the disorder to occur.

Age at onset of RLS symptoms provides the only established phenotypes.²⁵ Family history studies have repeatedly demonstrated increased familial occurrence of RLS for those RLS symptoms starting earlier in life (before age 45 in one study²⁶ and age 30 in another.²⁷ Thus, the genetic contribution appears to be greater for the early-onset than the late-onset RLS patients. The early-onset phenotype has also been found to have a relatively slow and persisting progression of the symptoms, whereas the late-onset phenotype shows a rapid progression to a stable level of severity,²⁵ usually occurring in less than 5 years and in some cases almost immediately with the onset of the symptoms. Studies have indicated that early-onset RLS (symptoms starting before age 45) appears to largely represent those whose RLS is caused by brain iron insufficiency, whereas late-onset RLS may be a mixture of causes, possibly including some with low brain iron levels.^28,²⁹ Despite apparent differences in underlying pathology, there are no known differences in treatment responses for these phenotypes.

RLS secondary to another medical condition creates another phenotypic distinction for RLS. RLS occurs commonly with three major medical conditions and with each usually resolves when the medical condition resolves (i.e., for end-stage renal disease, pregnancy, and iron deficiency). These are discussed in more detail in Chapters 25-28 in this book. Despite the differing pathologies, the presentation of symptoms and treatment responses are similar for all of these conditions except for the importance of treating the underlying condition.

Partial expression of RLS has been reported to occur. The PLM motor sign of RLS may occur more commonly in families with RLS than in those without any RLS. This has been demonstrated in one family study for older (over 50) subjects but remains to be explored in more detail.¹⁹ This may indicate that PLM occurring without involvement of another disorder may represent a forme fruste of RLS and may have essentially the same pathophysiology as RLS or, alternatively, there is a genetic factor contributing to PLM that also contributes to RLS.

In addition, a small set of cases has been reported with all the symptoms of RLS and involuntary leg movements but not the urge to move and in some cases neither the urge to move nor the abnormal sensations. These have been called quiescegenic nocturnal dyskinesias.³⁰ They probably do not involve the same pathology as RLS or may involve some overlap of pathology with RLS with added pathologic features producing this phenotypic variant of RLS.

Tools Aiding Diagnosis

Fortunately, some very good tools have been developed for making the diagnosis of RLS. Structured diagnostic interviews and patient-completed diagnostic questionnaires have been developed and adequately validated to be considered standards for the field. These also serve to provide information about phenotypic features of RLS. In addition, objective measures of leg movements using activity meters and physiologic recordings of leg movements in sleep and waking also provide strongly supportive information for the diagnosis.

Structured diagnostic interview. A structured diagnostic interview developed at Johns Hopkins explores both the basic diagnostic criteria and the differential diagnosis to exclude conditions that mimic RLS symptoms as noted earlier. The interview permits gradations of RLS diagnosis of definite, probable, possible, and not RLS. The occurrence of RLS symptoms primarily (≥90% of symptoms) when sitting changes an otherwise definite to a probable diagnosis. This reflects the uncertainty that any rest position engenders RLS, because lying down also produces rest. It may be, however, that for some, lying down without sleep does not last long enough to engender RLS symptoms, or that there is some conditioned arousal with lying down before sleep that disrupts expression of RLS symptoms—thus, the label of “probable RLS.” The lack of urge to move with the presence of all other features of RLS is in this interview schedule considered to be “possible RLS.” As noted, this diagnosis is most likely not RLS, and in many studies, patients with this diagnosis can be considered as not having RLS. The final diagnosis of “not RLS” applies only to those who clearly fail to have at least one of the four major diagnostic criteria for RLS. This interview has been adapted for a telephone diagnostic interview. When used by trained interviewers, the reliability between interviewers for this Hopkins Telephone Diagnostic Interview (HDTI) was found to be 95% in a population of RLS patients with non-RLS control subjects. Comparison with a standard clinical interview showed a sensitivity of 97% and specificity of 92%.³¹ In another study within a family case-control study, the HDTI had a specificity and a sensitivity of over 90% compared with paired clinical diagnostic interviews.³² This telephone interview, however, requires specific 1.5-day training offered by Johns Hopkins to be used with this degree of accuracy. It is unclear whether a similar degree of training is needed for the use of the structured diagnostic interview in a clinical setting.

The Restless Legs Syndrome Diagnostic Index (RLS-DI ³³). The RLS-DI consists of 10 items (Table 15-2). All essential criteria of the IRLSSG are included in the first five items (Part A). In addition, all features that are classified as “associated” with RLS by the IRLSSG⁸ were added (PLM in PSG or actigraphy assessment, response to dopaminergic therapy, RLS in family members). Two supportive features—sleep disturbances and findings of a neurologic assessment—complete Part B. Weights have been assigned to optimize accuracy of diagnosis and suggested degrees of confidence provided. A clinical diagnosis of RLS requires a total of 11 points as well as at least one item from 7 to 10 scored as 2.

A validation study was designed and conducted with the cohort of all 261 patients who were evaluated with the need to clarify any sleep-related disorder in one sleep laboratory.³⁴ The analysis of the four essential diagnostic criteria revealed a high sensitivity (Table 2, Part A: 94.1%) but a specificity that was too low (81.7%). That means that 18.3% of all patients with other diagnoses for their sleep disturbances met the essential diagnostic criteria (cutoff = 4 for Part A). The items of Part B improved significantly the specificity of the RLS diagnosis (p <.0001). The inspection of the ten diagnostic criteria revealed that both sensitivity and specificity depend on the strength of diagnostic requirements. Sensitivity of the criterion is high if a variable has to be present on at least 1 day per week (Part A) or was at least questionably present (Part B); specificity is high if an item was present on 5 to 7 days per week (Part A) or was definitely present (Part B).

Diagnostic questionnaire. A patient-completed diagnostic interview has also been developed (Table 15-4) in collaboration between the University of Cambridge, England, and Johns Hopkins University. This has been validated against the HTDI and found to have a positive predictive value of approximately 90%. This questionnaire covers not only the major diagnostic issues with RLS but also phenotype issues of age at onset, rate of progression, occurrence of “pain” with RLS, and occurrence of involuntary leg movements.

Diagnostic questions for surveys. Many epidemiologic studies rely on a small set of three to five questions to define those with RLS in the population studied. There have been two sets of these that have been validated. Berger and colleagues ³⁵ developed three questions for their diagnosis of RLS in a survey of adults in a northeastern area of Germany. Validation of these questions against direct clinical adjustment showed reasonable agreement (unweighted κ = 0.67).³⁶

A second set of five diagnostic questions was developed for use first in a small primary care setting and then applied in the two REST studies (Table 15-3).^37,³⁸ A later evaluation indicated this set could be reduced to four questions. The validation in the small primary care center showed good sensitivity and specificity of 82.3% and 89.4%, respectively.³⁹ Several of the more recent survey studies have used this set of questions for identification of RLS among the population studied.

Single screening question. A single screening question has been tested in several languages (Italian, Czech, Polish, and Japanese) with a published validation in Italian ⁴⁰ and an abstract reporting English results.⁴¹ The question reads:

TABLE 15-2 The Restless Legs Syndrome Diagnostic Index (RLS-DI)