Chapter 17 Periodic Limb Movements in Sleep
Periodic limb movements during sleep (PLMS) are involuntary, repetitive, stereotypic, short-lasting segmental movements of the lower and sometimes upper limbs. These are commonly described as consisting of a dorsiflexion of the big toe with fanning of the small toes, accompanied by flexion at the ankles, knees, and thighs (Fig, 17-1), although recent studies have indicated a wider variety of movement characteristics not limited to these classic physiologic flexor movements.1,2 Each movement lasts 0.5 to 10 seconds and occurs at intervals of 5 to 90 seconds, with a remarkable periodicity of approximately 20 to 40 seconds.3-6 Movements often involve both legs but may predominate in one leg or alternate between legs (Fig. 17-2). PLMS were first described by Symonds7 under the term “nocturnal myoclonus” (NM). Symonds reported a series of different motor phenomena occurring in sleep, which shared the common feature of muscular contractions of the extremities and, in the absence of polygraphic recordings, postulated an epileptic origin of the jerks. In the mid-1960s, Lugaresi and colleagues3 first recorded NM polygraphically in patients with restless legs syndrome (RLS) and other neurologic diseases and published a series of studies on its electroencephalographic and electromyelographic correlates. In 1980, Coleman and coworkers8 suggested the term “periodic movements in sleep” because the muscular contractions characterizing NM are not, in general, truly myoclonic and usually have muscle potentials longer than those characteristic of myoclonus (i.e., <250 milliseconds). The subsequent terms “periodic leg movements in sleep” and the most recent term “periodic limb movements in sleep” used by the International Classification of Sleep Disorders (ICSD)9 emphasize the observation that PLMS are usually present in the legs but that the arms may also be involved10 (Fig. 17-3).
That PLMS in a few patients may become pathological was recognized by the inclusion of the periodic limb movement disorder (PLMD) in the first International Classification of Sleep Disorders (ICSD) published in 1990.11 At that time, PLMD diagnosis required a PLMS index of five movements per hour or more, associated with an otherwise unexplained sleep-wake complaint, impacting either sleep continuity (insomnia) or daytime functioning (sleepiness) and in the absence of concurrent conditions or diseases accounting for the symptoms. Some clinicians criticize the validity and specific usefulness of this distinct diagnostic category, noting that, if it exists, it affects few patients. They suggest that patients must exhibit a clear perception of leg movements causing disruption of the patient’s or sleeping partner’s sleep for a diagnosis of PLMD.12 The update for Version 2 of the ICSD altered the diagnostic criteria for PLMD in three major ways. First, it suggested that for adults a frequency of greater than 15 per hour was required to be considered abnormal. Second, it emphasized excluding leg movements associated with end-of-sleep disordered breathing events. And third, it required PLMD severity judgment to be based on the clinical context rather than an absolute rate of PLMS.9
Epidemiology
Although most frequently associated with RLS, PLMS do occur in healthy people.13,14 PLMS may begin at any age, but in healthy adults they appear to be relatively uncommon until age 40 and then increase markedly with age.5,13,15 The prevalence of PLMS in all adults has been estimated at 5% to 6% in normal subjects13 and up to 18% in sleep disorder clinic populations.16–19 Forty-five percent of normal individuals over 65 may have a PLMS index of greater than 5,12 with an equal distribution among men and women. PLMS can show significant variability across nights,13,20–22 suggesting caution in drawing conclusions from single-night studies in individuals with mild forms of PLMS,23 especially because it seems to have little effect on PLMD severity classification or clinical treatment decisions.22
Normal children seem less likely to have PLMS.14,24,25 In a select population of children at increased risk for having a sleep disorder, the prevalence of isolated PLMS was only 1.2%.26 Limited data indicate that PLMD appear, however, to be common in children with restless legs syndrome (RLS), attention-deficit/hyperactivity disorder, oppositional disorders, and Williams syndrome.24,25,27–29 As in adults, PLMD and RLS in children are two separate but related entities. Children with PLMD and a family history of RLS are considered to be at risk for having or developing RLS.30
Polysomnographic Features
The presence of PLMS is assessed by polysomnographic (PSG) recordings using two surface electrodes placed over both anterior tibialis muscles according to standard criteria.31 Methods for recording and scoring PLMS currently accepted are based on the amplitude of anterior tibialis electromyography (EMG). Each event starts when the EMG amplitude exceeds 8 μV above baseline and ends when the amplitude remains below 2 μV above baseline for at least 0.5 second. The movements must be 0.5 to 10 seconds in duration. A sequence of 4 or more such movements during any sleep stage separated by an interval of at least 5 seconds and not more than 90 seconds is considered PLMS.6 Clinicians have often set a PLMS index of greater than 5 movements per hour of sleep or higher as significant, although this figure may be low, especially in the elderly.11,21 More recent standards set the criteria at greater than 15 movements per hour for adults while remaining at greater than 5 per hour for children.9 One study indicated that for healthy adults younger than 40 years, PLMS generally occurred at rates less than 5 per hour, but adults over age 40 showed rates rapidly increasing with age, with an average of 15 per hour in the 50- to 59-year-old age group and 20 per hour among those over age 60.15 On PSG recordings, the EMG activity often begins with a myoclonic jerk, followed, after a short interval, by a tonic contraction. In rare cases, the initial myoclonic jerk is lacking. More often, repeated myoclonic jerks occur at the beginning of each single movement (Fig. 17-4). Occasionally, repetitive, small-amplitude rhythmic activity may be found in the intervals between PLMS.32
The use of actigraphy for PLMS assessment has been reported,33 and actigraphic techniques34 have also been proposed as a practical and reliable tool for community studies and for automatic detection of PLMS.35,36 This method, however, awaits further validation before its use becomes widespread. Many clinicians are looking for novel, low-cost devices for PLMS testing at home.37
Periodic Limb Movement in Sleep Distribution During the Night and in the Different Sleep Stages
PLMS are most frequent during non–rapid eye movement (NREM) sleep stages 1 and 2. The movements become less frequent and the interval between movements becomes longer as sleep deepens along the NREM sleep stages.5,38 PLMS display a pattern of progressive decline alongside the exponential decline in slow wave activity throughout the night.39 PLMS are maximally suppressed during REM sleep, when the relative frequency is as low as during slow-wave sleep (SWS), movement duration is even shorter, and the intermovement intervals are longer5,38 (Fig. 17-5).
In patients in whom PLMS are found during sleep, PLMS of the same type may occur during quiet wakefulness, before sleep onset, or in the course of nocturnal waking episodes.3,38,40 The movements during wakefulness have been called periodic limb movements while awake (PLMW) or dyskinesia while awake (DWA). PLMW appear especially in patients with RLS, and, at least for this kind of patients, it is recommended to score PLM during both wakefulness and sleep to get a more complete picture of the motor disturbance.41
Two distinct patterns of PLMS activity have been identified.42 In the type 1 pattern, PLMS activity is initially high on falling asleep and then decreases across the remainder of the night with the majority of the activity occurring during the first half of the night; this distribution is typical of patients with either PLMS alone or PLMS associated with RLS. In patients with RLS, PLMS peaks between 11:00 P.M. and 4:00 a.m., showing an acrophase at approximately 3:00 a.m., while the lowest values are between 9:00 A.M. and 3:00 P.M.43–45 In the type 2 pattern, PLMS activity is relatively evenly distributed across the night with slightly greater activity in the middle of the night; this is typical of narcoleptic patients and patients with obstructive sleep apnea.42,46
PLMS may also vary with body position in sleep, with a significant tendency of a series of movements to terminate soon before body position changes and a trend for a series to begin soon after position changes.47
Periodic Limb Movements in Sleep and Arousal
PLMS may be associated with no changes in the electroencephalogram (EEG) and no other evidence of arousal or may instead lead to partial or full arousals (Fig. 17-6). EEG correlates of PLMS are K-complexes, K-alpha complexes, brief bursts of alpha activity, or other cortical and subcortical arousals.3,4,39 Arousals that are associated with PLMS (PLMS arousals) are calculated according to the arousal scoring rules of the American Sleep Disorders Association (ASDA).48
Microstructural analysis of EEG shows that in NREM sleep, the timing and location of arousal-related EEG transients are arrayed in a pseudoperiodic rhythm known as the cyclic alternating pattern (CAP).49 The CAP modulates several pathological phenomena and the great majority of PLMS are associated with phase A of the CAP.50
The sleep structure varies between those with PLMS alone and those with RLS. Patients with isolated PLMS in sleep have more frequent arousals associated with the PLMS than do patients with RLS and PLMS. However, patients with RLS spend significantly more time in wake-after-sleep-onset, have more arousals not associated with any sleep-related event, and more REM, but less stage 1 NREM, sleep than do those with isolated PLMS.51
Generally, PLMS not associated with visible EEG arousals or microarousals are considered to have little effect on sleep quality or daytime alertness,52,53 but a cause-effect relationship between number of arousals and daytime somnolence is not yet well established.5,54 Moreover, the time relationship of the EEG arousals with the leg movements varies: arousals or microarousals can follow the limb movements but can also precede or accompany them,55,56 suggesting that arousals are not simply the consequence of PLMS and that probably EEG arousals and PLMS are separate expressions of a common mechanism.57 The persistence of arousals after suppression of PLMS by L-DOPA41,55 and apomorphine58 supports the hypothesis that the EEG events are a distinct phenomenon, independent of PLMS and not related to the dopaminergic system. Therefore, the PLMS should be considered not simply a motor phenomenon occurring periodically and inducing sleep fragmentation but the expression of an underlying arousal disorder related to dysfunction of the oscillatory networks regulating the cyclic arousability of the sleepy brain.56
It is increasingly clear that central nervous system–orienting systems may be activated even without visible EEG changes. So PLMS may be associated with pure autonomic activation alone, in the absence of an EEG arousal as defined by the ASDA standards.59,60 Spectral EEG and heart rate (HR) analyses at the time of PLMS in patients with RLS and/or PLMD revealed a wide variety of complex and stereotyped variations in cortical activity and heart rate, associated with the PLMS. These included an increase in heart rate and delta band EEG activity before the PLM, which were independent of the presence of microarousals.61,62 The amplitude of the heart rate changes associated with PLMS varies with age.63 The same stereotyped pattern is common in PLMS and in spontaneous or stimuli-induced arousals.62 The concomitant cerebral and autonomic responses before and during PLM suggest that PLMS do not trigger cardiac and EEG activation but are parts of the entire arousal response, which recruits several neural networks from a common generator, probably located in the brainstem and projecting up to a cortical level.59,61,64 The different expression of arousal responses associated with the PLMS may finally mean that analysis of “cortical arousal” as proposed by ASDA criteria may not allow full insight into the real sleep fragmentation occurring in patients with PLMS.
Pathophysiology of Periodic Limb Movements in Sleep
The exact origin and pathophysiology of the PLMS are unknown. Although reduced cortical inhibition65,66 and impairment of cortical-subcortical motor structures, in particular motor inhibitory pathways67 have been reported in RLS, a cortical origin for the PLMS is unlikely, because electrophysiologic studies demonstrated the lack of any Bereitschaftspotential preceding the PLMS.32,68
It has been hypothesized that PLMS may result from rhythmical fluctuations of reticular excitability50,64 and that suprasegmental disinhibition during sleep may provoke PLMS.68–71 Neurophysiologic studies support the hypothesis that the underlying mechanisms are active at the pontine level or rostral to it.72 In PLMS patients, the excitability of the R2 response of the blink reflex was markedly enhanced.73
High-resolution functional magnetic resonance imaging (fMRI) in patients with PLMS and RLS showed significant activation of the red nucleus and the brainstem and no apparent involvement of the cortex.71
Because of their resemblance to the Babinski response and the normality of lower and upper extremity somatosensory evoked responses and brainstem auditory evoked responses, some investigators attribute the PLMS to suppression of supraspinal descending inhibitory pathways on the pyramidal tract.74,75
The observation that PLMS may be present below a complete spinal cord lesion,70,76–78 such as during midthoracic epidural or spinal anesthesia, indicates that PLMS can be directly generated in the spinal cord.79 Patients with PLMS and uremic RLS have increased spinal flexor reflexes and significantly increased spinal cord excitability particularly during sleep.80,81
On the other hand, the peripheral nervous system has been implicated on the basis of findings of small-fiber neuropathy in patients with primary or secondary RLS,32,82–86 but it is still an unresolved question whether the peripheral dysfunction triggers PLMS or is simply coincidental. Pacing of PLMS by repetitive electrical stimulation of the peroneal nerve32 indicates that peripheral sensory afferents play a role in triggering the PLMS. Although abnormal electromyographic activity of limb movements during wakefulness in RLS has been found to spread along propriospinal pathways intrinsic to the cord,68 no propriospinal pattern of propagation was found in two studies of PLMS showing that the leg muscles are the most frequently involved with no hierarchical order and no caudal or rostral propagation typical of propriospinal myoclonus.1,2 Muscle activation did not show a consistent recruitment pattern from one PLMS to another, indicating the engagement of different, independent, and sometimes unsynchronized generators for each PLMS (Fig. 17-7). These data suggested an abnormal hyperexcitability along the entire spinal cord, especially in its lumbosacral and cervical segments, as a primary cause of PLMS, triggered by state-dependent sleep-related factors located at a supraspinal level.1,2 On the other hand, axial jerks with a propriospinal type of propagation during relaxed wakefulness giving place to typical PLMS as soon as spindle activity shows up on the EEG recordings have been reported in RLS.87
The efficacy of dopaminergic drugs88,89 and opioids90–92 in reducing PLMS suggests an involvement of either the dopaminergic or the opioid system in their pathogenesis. In particular, the fact that PLMS are more frequent in RLS, narcolepsy, and most probably in REM sleep behavior disorder (RBD)93,94 raises the possibility of an impaired central dopaminergic transmission.94 The association between PLMS and Parkinson’s disease, although not as strong as one would have expected, also suggests that dopaminergic systems may be involved in the PLMS.95 Neuroimaging studies in RLS and PLMS, albeit controversial, disclose some evidence of an underlying dopaminergic pathomechanism of PLMS.96–101 Finally, in view of the role of iron in the origin of RLS, it was observed that patients with RLS and low serum ferritin have higher PLMS indexes.102
Periodic Limb Movements in Sleep (PLMS) in Sleep Disorders and Other Diseases
PLMS can occur as an isolated phenomenon in subjects without any sleep complaint4,103 or may be associated with a wide variety of medical, neurologic, and sleep disorders.8 PLMS were first polygraphically documented in the RLS,3 and more than 80% of patients with RLS experience PLMS.55 Elevated PLMS indices are found in up to 90% of narcoleptics over the age of 60 years.8,94 PLMS have also been reported in patients with obstructive sleep apnea syndrome (OSAS),104 who may have PLMS independent of the apneic episodes105 and in more subtle forms of sleep disordered breathing (i.e., the upper airway resistance syndrome).106 Depending on OSAS severity, PLMS may either increase or decrease following treatment of sleep apnea with continuous positive airway pressure (CPAP), which could unmask an underlying PLMD.107
PLMS are also highly prevalent in RBD,108,109 insomniac,3,8
