34 Essential myoclonus

The term ‘essential myoclonus’ refers to a disorder of unknown aetiology in which myoclonus is essentially the only neurological abnormality. Essential myoclonus appears to be very rare. There are some historical reports (see later); however, what exactly was wrong with the original patients is not clear. On the other hand, many patients with predominant myoclonus (which may have been called essential myoclonus) have been found to also have features of dystonia and this disorder has been classified as myoclonus dystonia which is in many (but not all) cases due to mutations of the epsilon sarcoglycan gene on chromosome 7. On the other hand, there is some confusion in the literature with juvenile and progressive myoclonic epilepsies, but again follow-up genetic data are in most cases not available for the original families.

Essential myoclonus is defined as an illness characterized by myoclonus, usually multifocal, for which no secondary cause can be found, without other neurological deficit, and often inherited as an autosomal dominant trait.

There are a relatively small number of reports of familial essential myoclonus in the historical literature. The original patient with ‘paramyoclonus multiplex’ described by Friedreich (1881) is taken as the first sporadic example of essential myoclonus. He exhibited multifocal, bilateral, asynchronous jerks of the arms and legs, and to a lesser extent of the face and trunk. The overall picture described was one of continuous chaotic muscular twitching varying in intensity and frequency from 10 to 50 per minute. The muscle jerks were said to be maximal at rest and disappeared during voluntary movement. There were no other signs of neurological disease, and there was no family history of any such condition. Biemond (1963) described a family using the same terminology of ‘paramyoclonus multiplex’.

Lindenmulder (1933) described familial benign myoclonus occurring in three successive generations. Mahloudji and Pikielny (1967) were the first to employ the term hereditary essential myoclonus. Daube and Peters (1966) described the laboratory and clinical findings in 12 patients from two families with hereditary essential myoclonus. The condition appeared to be inherited as an autosomal dominant trait. Chaotic, arrhythmic jerks of the arms, neck, trunk, and legs began between the ages of 1 and 20 years and persisted throughout life. The jerks were absent or reduced at rest but were increased on action. They were of relatively small amplitude and produced remarkably little disability over many years. Otherwise the patients were normal; there were no other neurological deficits, intellect was preserved, epilepsy did not occur, and the EEG was unremarkable.

Reviewing the literature, Mahloudji and Pikielny (1967) proposed diagnostic criteria for familial essential myoclonus (Table 34.1). They noted that the myoclonus was neither rhythmic nor synchronized. It was multifocal, with a predilection for the face, trunk, and the proximal limb muscles. The movements could affect a limited part of the body, be confined to one side, or affect the whole body. The jerks occurred spontaneously and the effect of movement was variable, increasing with action in several individuals, but decreasing in one. Stimulus-sensitivity was not noted. The myoclonus was absent during sleep and was aggravated by emotional distress. Korten et al. (1974) described another family in which 12 members were affected over two generations. This family was later found to have a heterozygous 2-bp deletion (619delAG) in exon 5 of the epsilon sarcoglycan gene (see later), resulting in a frameshift and premature truncation (Foncke et al. 2006). Their report was of particular interest for two reasons. First, there was a remarkable association between typical essential tremor and myoclonus in this family; four of the patients with myoclonus also had essential tremor and seven other members of the family had benign essential tremor only. The authors concluded that ‘there is an intrinsic relationship between essential tremor and essential myoclonus’. Their conclusion was supported by the observation that both the tremor and the myoclonus were improved by alcohol in this family; this had also been reported in another family with hereditary essential myoclonus described by Hallett et al. (1977). Second, the myoclonus in the Korten et al. (1974) family varied considerably from time-to-time; ‘periods of virtually rhythmic discharges alternated with phases of asynchronous, dysrhythmic myoclonus complexes, alternating with bilaterally synchronous complexes’.

Other families have been reported by Przuntek and Muhr (1983), Duvoisin (1984), Lundemo and Persson (1985),  Gualandri et al. (1987), Quinn et al. (1988), Fahn and Sjaastad (1991), Phanthumchinda (1991), and Alves et al. (1994). In the family described by Quinn (1988) the proband, a 25-year-old woman, developed symptoms at the age of 7 years. Two of her six siblings also were affected, as was their father. Clinical examination revealed pure myoclonus without dystonia, principally affecting the neck, arms, and trunk, with minimal jerks in the legs. The myoclonic jerks were sometimes synchronous and symmetrical, sometimes asynchronous and asymmetrical. They were not sensitive to stretch, touch, or light, but could be provoked by sudden loud noises. Speech was normal. Alcohol strikingly diminished the jerks, which were characteristically worse premenstrually and during her periods. In one large Norwegian family (Fig. 34.1) 11 out of 56 members were identified as having definite, probable, or possible myoclonus and a further two were observed to have involuntary movements under stress. Two others had a history consistent with myoclonus but had developed long-term remission (Fahn and Sjaastad 1991). Three members of this pedigree had features of mild focal dystonia and alcohol reduced the myoclonus. The myoclonus in this family was worse in a stress situation, which was also seen in a Thai family with inherited essential myoclonus. Alcohol also settled the myoclonus in this latter pedigree, but three out of 11 members also had generalized seizures (Phanthumchinda 1991).

Nutt and Bird (1984) reported a familial kindred in which essential myoclonus was associated with malignant melanoma.

There have also been a number of families reported, especially from Japan, in which there is onset of myoclonic jerks in upper and lower limbs, tremulous finger movements, and rare generalized tonic-clonic seizures (Uyama et al. 1985, Yasuda et al. 1991, Nagayama et al. 1998). Onset is in adult life, extensive investigation has not revealed any underlying cause, and other neurological features do not develop. The myoclonic episodes can be precipitated by fatigue, insomnia, and photic stimulation. The severity of the myoclonus may increase with age (Nagayama et al. 1998, Plaster et al. 1999). These families can be classified as having myoclonic epilepsy; however, as seizures are usually infrequent and do not necessarily occur in all members with myoclonus, they may be confused as having ‘essential’ myoclonus when in fact they do have a different disorder from the other cases mentioned above. This inherited form of essential myoclonus, which is associated with rare tonic-clonic seizures and is more common in Asians, may also be confused with the disorder(s) that have just been discussed. It is, however, quite a separate entity. Onset is later and it commences in adult life. There are also frequent, small, tremulous myoclonic movements of the fingers. This has been termed familial myoclonic epilepsy (FAME), cortical myoclonic tremor, and cortical tremor, and the gene has been localized to chromosome 8q24 and 2p11 (Okuma et al. 1998, Plaster et al. 1999) (also see later and Chapters 19, 29, and 30).

In addition to the small number of families described with inherited essential myoclonus, many sporadic cases have been reported. Only one patient amongst the 19 reviewed by Aigner and Mulder (1960) gave a family history of myoclonus and none of the patients with essential myoclonus identified by Caviness et al. (1999) had affected relatives. Bressman and Fahn (1986) reviewed their experience of essential myoclonus which they diagnosed on the following criteria: (1) myoclonus was the sole or primary neurological abnormality; and (2) there was no evidence of any preceding causal event or factor by history or laboratory examination. They did not exclude patients in whom there was mild ataxia (one case) or tremor (two cases). They included patients with rhythmic or segmental myoclonus if by history and examination there was nothing to suggest any other cause.

Aigner and Mulder (1960), in a follow-up study of 94 patients with myoclonus seen at the Mayo Clinic over a 12-year period, identified 19 cases of essential myoclonus and pointed out its benign nature. In a further study from the Mayo Clinic, Caviness et al. (1999) estimated the annual incidence of myoclonus to be 1.3 cases per 1000 person years and essential myoclonus made up 11% of these. The figures, however, may have been inaccurate due to the small numbers involved and case ascertainment through a records-linkage system. Bressman and Fahn (1986) reported 15 patients (Table 34.2). Their age at onset ranged from 2 to 64 years with a mean of 29 years. The duration of their illness varied from 1 to 19 years with a mean of 7 years. In no case was there marked progression of the illness or disability. All 15 patients were independent in daily activities and most led active lives. Only one of their cases gave a family history of myoclonus which was compatible with autosomal dominant transmission. One other patient gave a family history of tremor affecting his mother and a maternal aunt. All the patients had no other neurological abnormality on examination, except for the one with a mild limb and gait ataxia. The myoclonus most frequently involved the trunk and proximal limbs, then the neck and the face. Distal limbs were infrequently involved. In nine patients the myoclonus was focal or segmental. Involved regions included the right arm in one, both lower extremities in another, cranial nerve musculature in two, and the trunk in five. They noted that spread of myoclonus might be delayed for many years from the onset; one patient had neck involvement for 15 years before spread to the trunk. Six patients had diffuse or multifocal involvement. In 10 patients the myoclonus was synchronous but not absolutely constantly. The remaining five patients had myoclonus which was asynchronous, all with multifocal jerks. Two patients had rhythmic myoclonus, varying in frequency from 1 to 3 per second. Four patients had oscillatory myoclonus. The effect of action on the myoclonus was variable. Twelve patients had spontaneous jerks, which in two cases diminished when standing and walking. In three other patients the myoclonus was increased by specific motor actions. Sensitivity to peripheral stimuli was seen in the minority; sensitivity to stretch was noted in five patients, to painful stimuli in one, and to touch in another. Nine patients drank alcohol and five had benefit.

It will be seen that patients described as having familial or sporadic essential myoclonus form a heterogeneous collection. Three major problems emerge in assessing the significance of this entity:

Some families with an essential movement disorder have been described in which the precise nature of the movement has defied analysis. Movements have been described as myoclonic, choreic, dystonic, and tremulous (Lossos et al. 1997). The difficulty of arriving at a definitive diagnosis using clinical examination alone in patients with dominantly inherited jerks is well exemplified by the family described by Refsum and Sjaastad (1972) which was later found to have benign hereditary chorea. The prepositors presented generalized choreiform or myoclonic movements, mainly localized to the head, trunk, and proximal parts of the extremities. In a later article (Sjaastad et al. 1983), the authors commented ‘whereas the hyperkinetic events in the original description were interpreted as choreatic, we have been more inclined to interpret the movements as myoclonic’. Involuntary movements are most difficult to describe. Thus, there is also clinical overlap with benign inherited chorea (Haerer et al. 1966, Asmus et al. 2007a) (see Chapter 21).

Most of the patients with myoclonic dystonia (see Chapter 36) (but not all) have mutations in the epsilon sarcoglycane gene on chromosome 7q21. This condition is also known as dystonia with lightning jerks responsive to alcohol. Individuals within such families may exhibit either lightning myoclonic jerks alone or dystonia alone, or both. Psychiatric features (like obsessive behaviour and depression) are not uncommon. Indeed, Quinn et al. (1988) pointed out that many of the cases described under the title of hereditary essential myoclonus exhibited minor features of dystonia, as recorded in the original reports. Likewise, myoclonic dystonia and hereditary essential myoclonus may respond to alcohol. In a review of the topic, Quinn (1996) set out a number of features that have differentiated these two disorders in the literature (Table 34.3), but eventually came to the conclusion that they were probably the same. Now that the gene for hereditary myoclonic dystonia has been localized to chromosome 7q21-q31 (Nygaard et al. 1999, Zimprich et al. 2001), this situation should be clarified.

The epsilon sarcoglycan (SGCE) gene contains 12 exons and encodes the epsilon member of the sarcoglycan family, transmembrane components of the dystrophin–glycoprotein complex, which links the cytoskeleton to the extracellular matrix. Both mutations and deletions in the gene have been described. It is inherited in an autosomal dominant pattern; however, there is maternal imprinting (Grabowski et al. 2003), meaning that offspring receiving a mutant gene from their mother will almost never show symptoms, in contrast to those who receive a mutant gene from their father, where penetrance is almost complete.

With identification of the gene, the phenotype has been reassessed (Asmus et al. 2002, Doheny et al. 2002, Valente et al. 2005, Tezenas du Montcel et al. 2006).

In patients with large deletions (which may also affect neighbouring genes) the phenotype may be more complex (Asmus et al. 2007b). For example, Asmus et al. (2007) described three patients with heterozygous large deletions in the 7q21.13-21.3 region. However, the deletion size was variable and ranged from 1.63 to 8.78 Mb and up to 43 additional neighbouring genes were affected by the mutational change. Two of the patients presented with typical myoclonic dystonia, whereas one paediatric patient with split-hand/split-foot malformation and sensorineural hearing loss (SHFM1D, OMIM 220600) had not developed symptoms of myoclonic dystonia at the age of 9 years. This patient had the largest deletion of 8.78 Mb including the SHFM1, DLX6, and DLX5 gene.

The condition is discussed fully in Chapter 36.

By definition, routine investigation in patients with essential myoclonus is normal, including examination of the CSF and CT or MRI brain scan.

The results of neurophysiological investigation will depend on the type of essential myoclonus. In the variety that overlaps with or is identical to myoclonic dystonia with lightning jerks and response to alcohol, EMG studies based on a clinical diagnosis (prior to identification of the gene) have shown the duration of muscle bursts to be from 50 ms (Korten et al. 1974, Duvoisin 1984), through 60 ms (Sjaasted et al. 1983) to 120 ms (Biemond 1963). Hallett et al. (1977) describing the electrophysiological findings in two of three affected family members found that the EMG correlate of the myoclonic bursts consisted of triphasic agonist-antagonist-agonist bursts, each lasting 50–100 ms. These involuntary spontaneous EMG bursts were similar in form to the EMG pattern of normal voluntarily executed fast ballistic movements, so leading the authors to describe the physiological abnormality as ‘ballistic movement overflow myoclonus’. The other feature, leading to the term overflow was that when the patients jerked, prominent bursts of EMG activity of similar form and duration also occurred in many other muscles in the same or other limbs. In other words, there was an inappropriate spread of muscle activity to regions not normally activated by movement, causing the jerks (Fig. 34.2).

Following the gene identification, these findings were replicated in patients positive for the mutation in the epsilon sarcoglycan gene. Roze et al. (2008) observed a homogeneous electrophysiologic pattern of myoclonus of subcortical origin with short jerks (mean 95 ms, range 25–256 ms) at rest, during action, and during posture; there were no features of cortical hyperexcitability (specifically no abnormal C-reflex response and no short-latency premyoclonic potential on back-averaging studies). Somatosensory evoked potentials have been reported to be normal (Hallett et al. 1977, Duvoisin 1984, Shibasaki et al. 1986). Back-averaging showed no cortical correlate in the EEG preceding the EMG bursts in the patients described by Hallett et al. (1977) and Shibasaki et al. (1986).

However, in the familial patient described by Quinn et al. (1988) there was a generalized, symmetrical, and synchronous negative wave preceding the muscle jerks by 50 ms. Muscle biopsies are normal in mycolonic dystonia (Hjermind et al. 2008).

Three loci have been associated with this disorder which is also called benign adult familial myoclonic epilepsy or BAFME. One form, BAFME1, has been mapped to chromosome 8q24 (Mikami et al. 1999). Another type, BAFME2, has been mapped to chromosome 2p11.1-q12.2 (Guerrini et al. 2001). A third locus, FAME3, is associated with a more severe phenotype (Carr et al. 2007).

Familial adult myoclonic epilepsy (FAME) shows generalized spikes or multi-spikes and slow wave complexes on ECG with photosensitivity in many patients. Somatosensory evoked potentials are enlarged, C-reflexes are enhanced, and cortical spikes are demonstrable on back-averaged EEG triggered by the myoclonic jerk (Fig. 34.3). The neurological features are thus those of cortical reflex myoclonus (Terada et al. 1997, Plaster et al. 1999) (see also Chapters 19, 29, and 30).

In the large Italian kindred with BAFME2 (Guerrini et al. 2001) in which 11 individuals over five generations were affected with an autosomal dominant disorder characterized by distal myoclonus and seizures, all affected members had onset in adulthood (mean 25 years) of distal, rhythmic, involuntary movements resembling tremors, and infrequent generalized tonic-clonic seizures (GTCS). Three patients also had intractable complex partial seizures, which were often followed by secondary generalization. EEG revealed focal frontotemporal as well as generalized interictal abnormalities. Detailed neurophysiologic studies showed giant somatosensory evoked potentials, enhanced long-loop C-reflexes, and premovement cortical spikes by the jerk-locked averaging method, suggesting a cortical origin. The authors concluded that affected patients had diffuse cortical hyperexcitability and a high propensity for intra- and interhemispheric cortical spread. Neuropsychologic testing revealed mild to moderate mental retardation in three members. Several patients had low or low-normal intellectual functioning.

Some patients may have little disability and others are reluctant to take drugs. For others the myoclonic component of their movement disorder may be more incapacitating.

The beneficial effects of alcohol have been noted, but are not universal and should certainly not replace medical treatment. Clonazepam seems to have given greatest benefit (in a dose of 1–20 mg/day). In the series reported by Bressman and Fahn (1986) clonazepam was given to 13 patients, 10 of whom responded, sometimes to a considerable degree. Sodium valproate helped only one of seven patients so treated. Drugs reported to be of little benefit by Bressman and Fahn (1986) were methysergide, cyproheptadine, trasadone, reserpine, tetrabenazine, baclofen, clonidine, carbamazepine, propranolol, D-trytophan, and L-5-hydroxytrytophan. Korten et al. (1974), however, did record some benefit from propranolol. Clonazepam has been reported to improve myoclonus, but in families in which there has been associated dystonia (Alves et al. 1994) and in those in which there is associated epilepsy it may not be fully effective (Phanthumchinda 1991, Nagayama et al. 1998, Plaster et al. 1999). Anticholinerigcs may help associated dystonia in the former group (Quinn 1996), whereas valproate has been reported to be effective in the latter (Plaster et al. 1999). Stimulation of the ventral intermediate thalamic nucleus has been found to ameliorate drug-resistant myoclonus in inherited essential myoclonus associated with mild dystonia (Kupsch et al. 1999).

Occasionally small jerks occur in normal people if they are stressed or following exercise. These are a variety of physiological myoclonus (Marsden et al. 1982). Benign myoclonus of infancy starts between 3 and 8 months of age. It consists of repetitive flexor spasms which are unaccompanied by EEG or mental changes. It is benign and eventually disappears (Lombroso and Ferjerman 1977).

Amongst the parasomnias (events around sleep) (Table 34.4) are three types of nocturnal movements which have at different

times been called myoclonus-hypnic jerks, excessive fragmentary myoclonus, and periodic movements of sleep, all of which may be considered to be physiological events associated with sleep or pre-sleep. Furthermore, in infants there is the condition of benign sleep myoclonus of infancy. Sometimes they are a cause for concern, particularly if they awaken the sleeping partner. Early clinical descriptions of ‘nocturnal myoclonus’ (Symonds 1953, Oswald 1959) have been expanded by detailed observation and electrophysiological monitoring in ‘sleep laboratories’ (Guilleminault 1982, Parkes 1982, 1985 and 1986). These movements have been largely covered in Chapter 47. They range from being brief to quite prolonged, and in many cases they do not really fulfill the criteria for myoclonus (Montagna et al. 1988).

Normal humans may suddenly jerk while falling asleep (Oswald 1959). Everyone is familiar with the sudden massive whole-body start that suddenly awakens us as we drift off and this occurs in persons of all sexes and all ages with a prevalence of 70% (Walters 2007). They are thus also known as sleep starts. Sometimes such awakening is accompanied by a vivid fleeting dream. Many people, however, are not aware of these hypnic jerks. They are intensified by high pre-sleep arousal, and are most frequent after excessive emotional excitement or evening coffee drinking. Hypnic jerks rarely pose a problem, but occasionally if severe they may prevent sleep onset.

It is hypothesized that the jerks arise from sudden descending volleys that originate in the brainstem reticular formation and are activated by the instability of the system at the transition between wake and sleep (Walters 2007).

Historically, Symonds (1953) proposed that in those who experience such jerks ‘greatly in excess of those which may occur in normal persons’, the term nocturnal myoclonus should be used. He even wondered if they might be epileptic in origin, but Oswald (1959) considered them to be a normal phenomenon. Amongst 50 of his acquaintances he found only seven who had never recalled such hypnic jerks, 25 who had them up to three times a year, 13 who had them once every 1–2 months, and five who had them weekly. Oswald (1959) considered that they might represent reflex startle reactions, and noted that they sometimes occurred during small K-complexes in the EEG. Hypnic jerks are more common in those with markedly disturbed sleep behaviour (Ohayon et al. 1997).

The current diagnostic criteria as implemented in the International Classification of Sleep Disorders comprise: complaints of sudden brief jerks at sleep onset, mainly affecting the legs or arms; that jerks are associated with at least one condition from among a subjective feeling of falling, a sensory flash, or a hypnagogic dream; and that the disorder is not better explained by another sleep disorder, medical or neurologic disorder, mental disorder, medication use, or substance use disorder (Walters 2007).

The differential diagnosis of hypnic jerks includes restless leg syndrome (RLS) in which periodic and aperiodic involuntary movements are relatively common (see Walters 2007). In some patients with RLS (a minority though) the jerky movements movements may be more prominent than the leg discomfort.

Treatment is only rarely needed. Reassurance that this is a normal phenomenon is often enough. In any case, there are no known treatments.

This is familiar as the multifocal twitching of the limbs, face, and trunk of the dog asleep on the floor. In man (it is more common in older men), fragmentary physiological sleep jerks are usually of small amplitude and are limited to the hands, toes, and face (De Lisi 1932). In some patients the disorder is diagnosed only as an incidental finding on polysomnography and no visible movement is present (Walters 2007). They occur during non-REM stages I and II sleep and during REM sleep. Occasionally, these multifocal jerks may be excessive in severity and occur throughout non-REM sleep, even in stages III and IV (Broughton and Tolentino 1984, Broughton et al. 1985). For a formal diagnosis, the following criteria are required: the patient exhibits small movements of the fingers, toes, or corners of the mouth, or small muscle twitches resembling either physiologic hypnic myoclonus or fasciculations; the movements may be present during wakefulness or sleep; polysomnographic monitoring demonstrates recurrent and persistent very brief (75–150 ms) EMG potentials in various muscles occurring asynchronously and asymmetrically in a sustained manner without clustering (the EMG thus resembles the phasic REM twitches normally found in REM sleep, except that they exist in all stages of sleep and are not clustered as in normal REM sleep, but are more evenly spaced across individual epochs); more than five potentials per minute are sustained for at least 20 min of non-REM sleep stages 2, 3, or 4; and the disorder is not better explained by another sleep disorder, medical or neurologic disorder, medication use, or substance use disorder (see American Academy of Sleep Medicine 2005, Walters 2007).

The movements usually disappear spontaneously over the first 3–4 months and have been confused with epileptic seizures. The EEG is normal (Fig. 34.4) and the movements are not influenced by anticonvulsant medication (Resnick et al. 1986, Daoust-Roy and Seshia 1992, Di Capua et al. 1993). Treatment is usually not needed.

Benign sleep myoclonus of infancy is characterized by onset in the neonatal period and the fact that is does not persist beyond infancy. Movements occur only in sleep and stop the minute the child awakes. Sometimes the movements can be precipitated by rocking during sleep. These features differentiate it from sleep-related epilepsy; however, in cases of doubt, a EEG is helpful. Prevalence and aetiology remain unknown. The diagnostic criteria of this condition include, as based on the International Classification of Sleep Disorders (2005): repetitive myoclonic jerks involving the whole body, trunk, or limbs; the movements occur in early infancy, typically from birth to 6 months of age; the movements occur only during sleep; the movements stop abruptly and consistently when the infant is aroused; and the disorder is not better explained by another sleep disorder, medical or neurologic disorder, or medication use. No treatment is usually needed as the course is benign without sequelae and as a rule the symptoms disappear after a few months of life [see Walters (2007) for review of sleep-related movement disorders].

In contrast to whole-body hypnic jerks in pre-sleep and fragmentary sleep myoclonus during sleep, periodic movements are not brief jerks (Coleman et al. 1980, Lugaresi et al. 1986). They are rhythmic spasms, each lasting 3–5 seconds, often occurring in bursts or clusters starting with several clonic contractions followed by a tonic spasm, at intervals of 15–90 seconds or so, lasting for several minutes up to half an hour. The movements may affect one or both legs. The big toe and foot dorsiflexes and the other toes fan; sometimes the knee and hip flex, so that the whole leg bends in a flexor spasm (Smith 1985). Such periodic movements arise out of non-REM stage I and II sleep.

A family history can be obtained in up to one-third of cases. Periodic movements of sleep are a common, age-related phenomenon. While the prevalence of PLMS is estimated to be 4–11% in adults, PLMS occur rarely in children, although medical conditions like sleep apnoea syndrome or neuropsychiatric disorders can lead to increased rates. In the elderly, PLMS are also common in subjects without sleep disturbances. The incidence rises to about 30% of those over 50 years old (Lugaresi et al. 1986). The movements rarely disturb the patient but may awaken the sleeping partner. Rarely, similar leg movements, uni- or bilateral, may occur in pre-sleep and be recalled. More often, the patient complains of RLS, with which periodic movements of sleep is associated (see Chapter 47). Most patients with RLS have periodic movements of sleep, but most patients with the latter do not complain of restless legs. PLMS also often occur in narcolepsy, sleep apnoea syndrome, and REM sleep behaviour disorder. PLMS were found also in various medical and neurological disorders that do not primarily affect sleep.

The pattern of muscle recruitment and spatial spread resembles the spinal flexor reflex produced by electrical stimulation of the medial plantar nerve. Such flexor reflexes are more easily elicited in patients with RLS and periodic movement of sleep, even when they are sleeping, supporting the proposition that these activities are really a form of spinal flexor reflex (Bara-Jimenez et al. 2000). The appearance of periodic movements of sleep in those with spinal cord lesions and their relief following surgical removal supports this notion (Lee et al. 1996).

Dopaminergic agents such as levodopa/dopa decarboxylase inhibitors and dopamine agonists are considered the treatment of choice for PLMS as well as for RLS [for review of PLMS see Stiasny et al. (2002)].

Hypnic jerks, fragmentary sleep myoclonus, and periodic movements of sleep must be distinguished from other nocturnal motor events.

Bruxism (teeth grinding) occurs to a small degree in most people but is severe in about 0.2% of adults, 0.5–3% of infants, and the elderly and may then result in damage or a temperomandibular joint syndrome. In contrast, sleep-related faciomandibular myoclonic movements consist of rapid jaw jerks or twitches, compared to the more sustained jaw closure seen in bruxism (Aguglia et al. 1991, Walters 2007) (see Chapter 29).

Head banging, sometimes with body-rocking, is mainly confined to infants and children. Sleepwalking occurs in up to 15% of children between the ages of 5 and 12. It occurs during the first 1–2 hours of the night, during stage III or IV non-REM sleep. Actually, most children do not walk but sit up and make repetitive and purposeless movements, apparently unaware of their surroundings. If awoken, they are temporarily disorientated. Night terrors and partial arousals from non-REM stage III or IV sleep also occur in the first third of the night. The child awakes with a piercing scream, followed by anxiety, overactivity, and automatic behaviour with autonomic arousal without full alertness. REM sleep behaviour disorders affect the elderly (Schenck et al. 1987). They result from a loss of the normal muscle atonia of REM sleep, and occur therefore principally in the early morning. They usually involve vocalization and violent or aggressive actions that might reflect ‘the acting out of dreams’. Violent behaviour during sleep has been reported to occur in about 2% of adults and in such people other sleep disturbances, including talking, bruxism, hypnic jerks, night terrors, and hypnogogic hallucinations are more common than in the general population (Ohayan et al. 1997).

Most forms of epilepsy are common during sleep, especially during arousal in the morning. Generalized tonic-clonic seizures and partial epilepsies, especially those arising from the frontal lobes, have a propensity to occur in sleep. Frontal epilepsy often contains major complex motor manifestations. Paroxysmal hypnic dystonia probably is an expression of covert medial frontal seizure discharge (see Chapter 48).

Video-EEG telemetry during sleep can be very helpful in distinguishing the cause of obscure nocturnal episodes.

A hiccup consists of a sudden contraction of the inspiratory muscles, terminated by abrupt closure of the glottis, producing the characteristic sound. We all have them at one time or another, often precipitated by over-distention of the stomach, excitement, or alcohol (Lewis 1985). The function, if any, of a hiccup is unknown.

‘Hiccough’ is incorrect, for hiccup (singultus) is not a respiratory reflex. Bailey (1943) described a ‘hiccup reflex centre’ in the upper cervical segments of the spinal cord, between the third and fifth cervical segments, with afferent input from vagal sympathetics and sensory fibres of the phrenic nerve (Salem et al. 1967). The efferent limb is via the phrenic nerve to the diaphragm and nerves to the glottis and accessory muscles of respiration. Hiccups can persist even after transection of both phrenic nerves (Campbell 1940). Newsom-Davis (1970) concluded that hiccup was due to single or repeated bursts of activity in a supraspinal mechanism independent of the pathways controlling rhythmic breathing.

Hiccups occur at a frequency of 4–60 times a minute, usually in bouts. Bilateral or unilateral involvement of the diaphragm may occur. They may continue, or start and stop during sleep. Benign self-limited hiccups may be due to many causes (Table 34.5).

Persistent or intractable hiccups may be due to a variety of conditions (Table 34.6). The dubious honour for hiccuping the longest is said by the Guinness Book of Records (1984) to belong to Charles Osborn from Iowa who hiccuped some 420 million times for more than 60 years without any known cause!

Persistent hiccups occur more frequently in men than women. Amongst 220 such patients seen at the Mayo Clinic between 1935 and 1963, 181 were men and only 39 were women. Organic causes were found in 93% of men but only in 8% of women (Souadjian and Cain 1968).

The cause of persistent hiccups is usually located in the gastrointestinal tract, with gastro-oesophogeal reflux being the most frequent (Federspil and Zenk 1999). Local irritation of the diaphragm or vagus nerves in the abdomen or thorax by a large liver, tumour, or abscess is another frequent stimulus. It has been described as a sign of myocardial ischaemia. General anaesthesia and intra-abdominal surgery may also be a trigger. Hiccups may be a manifestation of uraemia, diabetes mellitus, or other metabolic derangements.

Medications may also cause hiccups. Corticosteroids appear to be the most common offending agent, followed by antidepressants, dopaminergic agents, antibiotics, digoxin, opiates, and non-steroidal anti-inflammatory drugs (Bagheri et al. 1999). Benzodiazepines have also been reported to be frequent amongst the offending agents (Thompson and Landry 1997).

Intrinsic cervical cord or medullary lesions may cause persistent hiccups. These may be due to tumour, infarction, arteriovenous malformation, or multiple sclerosis (Chang et al. 1994, Funakawa and Terao 1998, Musumeci et al. 2000). Hiccup often with vomiting may occur in raised intracranial pressure, meningitis, or encephalitis. Of particular interest were the epidemics of hiccups recorded during the outbreak of encephalitis lethargica in 1919–1924 (Brain 1923, Rosenow 1926).

Chronic hiccup may persist during the early stages of sleep but becomes progressively less marked when passing from stage I to stage IV. It may, however, disappear at sleep onset and is even more inclined to stop during REM sleep (Arnulf et al. 1996).

Many cures have been proposed for hiccups, including pulling on the tongue, sipping iced water, drinking from the far side of a glass, inhaling or swallowing noxious irritants, manual stimulation of the nasopharynx, breath holding, a thump on the back, inducing vomiting, rebreathing from a bag, breathing carbon dioxide, acupuncture, financial reward, and prayer (Lewis 1985). In those with severe intractable hiccups, many drugs have been claimed to halt a bout, including chlorpromazine (Davignon et al. 1955, Friedman 1996) and other dopamine antagonists such as haloperidol and metoclopramide (Williamson and Macintyre 1977), diphenylhydantoin (Newsom-Davis 1974), carbamazepine (McFarling and Susac 1974), sodium valproate (Jacobsen et al. 1981), nifedipine (Mukhopadhyay et al. 1986, Lipps et al. 1990), amitryptiline (Stalnikowicz et al. 1986), and methylphenidate (Vasiloff et al. 1965).

Baclofen, however, is perhaps the most effective agent and has been said to produce long-term complete resolution in 49% of cases of chronic hiccup (average duration of hiccup before treatment being 4.6 years) with improvement in a further 27% (Burke and White 1988, Guelaud et al. 1995). It is thought to reduce the excitability of the hiccup reflex. If there is gastric reflux, cisapride and omeprazole may also be effective and can be combined with baclofen (Petroianu et al. 1997). Gabapentin has been substituted for baclofen or used as add-on therapy in resistant cases (Petroianu et al. 2000). Attempts at surgical therapy have occasionally been undertaken in an attempt to interrupt a reflex arc (Federspil and Zenk 1999). Implantable breathing pacemakers, designed to control the excursion of the diaphragm by electrical stimulation of the phrenic nerve, have also been reported to be effective (Dobelle 1999).

The topic of diaphragmatic myoclonus, which may present as hiccup, is discussed in Chapter 29.