Article Text

Download PDFPDF

Intermittent head drops: the differential spectrum
  1. Elena Antelmi1,2,
  2. Giuseppe Plazzi1,3,
  3. Roberto Erro2,
  4. Paolo Tinuper1,3,
  5. Bettina Balint2,
  6. Rocco Liguori1,3,
  7. Kailash P Bhatia2
  1. 1Department of Biomedical and Neuromotor Sciences, Alma Mater Studiorum, University of Bologna, Bologna, Italy
  2. 2Sobell Department of Motor Neuroscience and Movement Disorders, University College London (UCL) Institute of Neurology, London, UK
  3. 3IRCCS Institute of Neurological Sciences, Bologna, Italy
  1. Correspondence to Dr Elena Antelmi, Sobell Department of Motor Neuroscience and Movement Disorders, University College London (UCL) Institute of Neurology, 33 Queen Square, London WC1N 3BG, UK; elenaantelmi{at}


Intermittent Head Drops are episodic head flexion movements that can occur in a number of conditions. Typically, the term has mainly been related to epileptic episodes, but the spectrum of clinical conditions associated with this feature is wide-ranging even if never discussed in detail. By searching the electronic database, we may find that apart from the epileptic conditions, Intermittent Head Drops have been in fact reported in the setting of movement disorders, sleep disorders and even internal medicine disorders, such as Sandifer syndrome. We render an in-depth description of this characteristic phenomenon in different diseases, describing the clinical clues and neurophysiological patterns that may help the clinician to distinguish between the different settings of occurrence.

View Full Text

Statistics from


Intermittent Head Drops (IHDs) are referred to as abrupt and repetitive episodes of head flexion occurring from one to many times a day in a regular or irregular fashion, and can be seen in a large spectrum of conditions. The term mainly cross-refers to epileptic episodes, namely atonic epileptic seizures, infantile spasms or negative myoclonus of the neck. However, IHDs may not be epileptic in nature. For instance, IHDs have been reported in chorea-acanthocytosis (CA),1 Huntington disease (HD),2 and in childhood and adulthood Narcolepsy type 1 (formerly known as narcolepsy with cataplexy).3–5

The broad spectrum of conditions associated with IHDs highlights the need for a rational diagnostic work up, since the management and the prognosis of such disorders differ considerably. We offer a comprehensive overview of IHDs, aiming to help clinicians to distinguish between different—sometimes treatable—disorders. We further attempted to render an in-depth phenomenological description of IHDs, characterising the peculiar features of this phenomenon in different disorders.

Search strategy and selection criteria

We searched the Medline database (via PubMed: for publications between 1980 and February 2015, using the search terms: head drops, head drop attacks, atonic attacks, paroxysmal head drop, epileptic attacks, myoclonus of the neck, infantile spasms, paroxysmal dyskinesias, head banging, head rolling, head nodding, head shaking. Only papers in English were selected and reference lists of the articles retrieved by the electronic searches were checked for other relevant reports not indexed in the electronic database. We excluded articles focusing on constant (ie, non-intermittent) head drops as seen in neuromuscular disorders (ie, amyotrophic lateral sclerosis, inflammatory myopathy, chronic inflammatory demyelinating polyneuropathy, hypothyroidism, congenital myopathy, myasthenia gravis), parkinsonian conditions (eg, multiple system atrophy), or in association with cervical spinal and plexus lesions.

The reference lists of the retrieved papers were checked to include original articles focusing on rare diseases, even if published before 1980.

IHDs may occur in epileptic syndromes, movement disorders (MDs) and sleep disorders. Accordingly, we will separately describe IHDs in Epileptic Syndromes, IHDs in Movement Disorders, IHDs in Sleep Disorders and, finally IHDs in clinical settings of uncertain classification (table 1).

Table 1

Details of clinical and neurophysiological features of IHDs in the reviewed disorders

We provide a video showing examples of IHDs in different disorders. Signed patient consent has been obtained for videos of any recognisable patient.


IHDs in epilepsy

The first phenomenological description of epileptic IHDs was by Lennox, who, in 1960,6 distinguished between epileptic phenomena of head drop and epileptic drop attacks leading to a complete loss of upright posture, so that he referred to the former as ‘the minimal form of astatic epilepsy’.

IHDs have been subsequently described in the spectrum of myoclonic-astatic epilepsy.7 In such cases, IHDs are due to cortical myoclonus involving neck muscles. Patients usually have other types of seizure and EEG frequently shows interictal abnormalities.8 ,9 They are often associated with neurodevelopment abnormalities.

Seizures presenting the semiology of IHDs have also been described in the spectrum of myoclonic epilepsy of infancy (MEI), and particularly in the atonic variant of the Benign childhood Epilepsy with Centro-Temporal Spikes (BECTS).10 In these patients, IHDs are characterised by abrupt and brief atonic attacks of the neck muscles, which are frequently repetitive and are associated with loss of muscle tone at the electromyography (EMG). Back-averaging detects EEG epileptic activity coupled with the EMG atonia.

Negative IHDs can also occur in neurometabolic disorders such as mitochondriocytopathies, storage disorders, neuronal ceroid lipofuscinosis, non-ketotic hyperglycinaemia, infantile hexosaminidase deficiency and biopterin deficiency.11 However, in such syndromes, myoclonus tends to be generalised and IHDs do not represent a predominant feature. Moreover, neurometabolic disorders are usually accompanied by developmental arrest or regression.

Finally, IHDs can occur as a consequence of antiepileptic drugs (mainly carbamazepine and oxcarbamazepine), or pathological metabolic conditions, such as hepatic or renal encephalopathies. However, in these conditions, myoclonic jerks rarely involve the proximal muscles and mainly manifest as an asymmetric and irregular flapping of the hands while the patient is holding the position of dorsiflexion of the wrists (asterixis).12

On the other hand, epileptic IHDs can be the result of an active phenomenon rather than from loss of muscular tone. Positive IHDs can, in fact, be a feature of myoclonic epilepsy, mainly of the mildest form of MEI and of reflex MEI.13 ,14

Further, positive IHDs may be the expression of infantile spasm of West syndrome or of other epileptic encephalopathies.11 ,15 Such spasms are subtle and brief, resulting in a head nod or tonic eye rolling. Spasms tend to cluster for a few minutes once every 10–20 min, and are characterised on EMG by a muscular contraction lasting from 1 to 2 s, with a characteristic ‘rhomboid shape’. The ictal EEG counterpart shows high-amplitude slow wave, with an inverse phase reversal over the vertex regions, often in the context of a dramatic hypsarrhythmic background. The clinical picture in these patients is usually more complex and characteristic, allowing proper recognition16 ,17 (see online supplementary video 1: segment 1).

IHDs in MDs

IHDs in tic disorders and stereotypies

Sudden head movements including IHDs are common in tic disorders, including Tourette syndrome. In such cases, head flexion or extension can represent a simple or a complex motor tic. They are characterised by abrupt onset and are usually brief and repetitive, but not rhythmic. In such cases, IHDs are the result of a tonic contraction, although occasionally the high velocity of the tics may mimic atonic drops (see online supplementary video 1: segment 2). A distinctive characteristic of tics is the possible presence of suppressibility and urge to perform the movements prior to the onset of the tics.18

Phenomenological similarities bridge tics to stereotypies,19 although the lack of urgency to perform the movements, and their suppression by distraction, can allow distinction between stereotypies and tics.20 Head stereotypies are quite common. An example is ‘head banging’ or ‘head nodding’ described in mentally disabled children,21 or in amphetamine poisoning.22 It consists of repetitive head nodding, presenting as side-to-side or up-and-down movements, with a frequency of 1–2/s. Movements can be stopped voluntarily, and may be associated with up-gaze eye deviation or motor activity in the hands or feet (see online supplementary video 1: segment 3).

IHDs in choreic syndromes

Schneider et al,1 described characteristic flexion and extension spasms of the neck and the trunk in four patients with advanced CA. Movements were described as sudden and violent flexion, or less frequently, extension of the head. Movements are dystonic and/or choreic in nature, even if they bear resemblance to ballistic movements (see online supplementary video 1: segment 4). IHDs in these patients can also share some similarities with negative myoclonus and tics. However, they are less ‘shock-like’ and repetitive than myoclonus and cannot be suppressed as tics. While Schneider et al claimed that IHDs are almost pathognomonic of CA, Morgante et al2 recently reported that IHDs can also occur in HD. In HD, the electrophysiological pattern of IHDs is characterised by EMG bursts of variable duration (mean duration >300 ms) without any EEG correlates, thus confirming their choreic nature. However, we note that IHDs in HD are generally more subtle and less violent than those observed in patients with CA (see online suppplementary video 1: segment 5). The suggestion that IHDs can occur in the other choreic syndromes has been subsequently confirmed by others investigators,23–26 who suggested that IHDs are a late clinical feature and a marker of severity of choreic syndromes.

IHDs in paroxysmal dyskinesias and other MDs

Paroxoysmal dyskinesias are a rare group of conditions, mainly hereditary, characterised by sudden involuntary movements of brief duration that recur episodically.27 ,28

Isolated paroxysmal dyskinesia of the neck, thus resembling IHDs, are quite uncommon except for a condition formerly known as benign paroxysmal torticollis of infancy (BPTI), which has been associated with the CACNA1A29 and PRRT2 mutation.30 ,31

BPTI is a self-limiting, age-dependent, benign disorder consisting of recurrent episodes of head tilt. The attacks occur spontaneously or are triggered by sudden movements and usually settle within a few minutes/hours, however, they can sometimes last longer (days). The attacks typically start during the first year of life and end by the age of 2–4 years. During the attacks, infants may cry and appear agitated, with abnormal axial posture and gait disturbance. The phenomenology of the movement may be more ‘wriggling’ and ‘twisting’ in nature and, accordingly, the EMG shows a dystonic pattern.32 A family history of similar episodes or of migraine is common. Neuroimaging studies are mandatory to rule out symptomatic forms (ie, posterior fossa lesion).

Bobble-head doll syndrome is, also, a rare but characteristic disorder described in children.33 ,34 It is characterised by episodic, involuntary head movements in flexion, extension or, less frequently, from side-to-side, which occur at a frequency of 2–3 Hz. Most of the reported cases are associated with communicating hydrocephalus due to lesions in the region of the third ventricle, so that diversion of cerebrospinal fluid by shunting prompted a dramatic improvement.

IHDs in sleep disorders

IHDs in sleep-related rhythmic MDs

Head stereotypies may occur during drowsiness or sleep and present some phenotypic overlap with those described in mentally disabled children. Head banging, indeed, is a type of sleep-related rhythmic MD (RMD)35 ,36 that affects infants and children, and involves large muscle groups (especially neck and trunk muscles). It results in repetitive, stereotyped and rhythmic movements (lifting of the head, banging forwards or backwards), emerging predominantly during drowsiness or sleep, at a frequency of 0.5–2 s, and usually occurring in clusters lasting less than 15 min. Diagnosis also requires a history of sleep interference, impaired daytime function and/or self-inflicted bodily injury.

Headbanging can be combined with other forms of RMD, such as body rocking.

IHDs in narcolepsy type 1

Narcolepsy type 1 is a rare life-long disorder characterised by daytime sleepiness, cataplexy, rapid transitions from wakefulness to rapid eye movement (REM) sleep and disturbed nocturnal sleep because of awakenings, and altered motor control due to loss of the posterior hypothalamic neuron-producing hypocretin (or orexin).35

Cataplectic attacks are typically characterised by sudden and brief loss of muscle tone with preserved consciousness; they are triggered by strong emotions (laughter or pleasant surprise) and are due to REM sleep muscle atonia occurring during wakefulness. During these events, EEG shows a normal wakefulness pattern, REMs can be seen on simultaneous oculogram, while the EMG reveals the waxing and waning loss of muscle tone.37 Ictal frontal lobe hypersynchronous paroxysmal theta bursts38 have been reported in Narcolepsy type 1 children during these attacks. Tendon reflexes are absent during cataplexy episodes and the H-reflex is inhibited or absent.39–41 Respiratory and extraocular muscles remain unaffected during the attacks.

Even if cataplexy often results in a complete paralysis of voluntary muscles and leads to falls, loss of tone can be partial, affecting only some muscles.3 ,37 In some cases, there is prominent facial involvement (described as ‘cataplectic facies’) and head drops.3 ,5 Frequently, there are superimposed active motor phenomena mainly involving the facial muscles, which can be confused with tics or dystonia.4

Video-polygraphic recordings show that atonia first involves neck muscles in more than 80% of cases,37 and eventually spreads to axial and lower limb muscles. Less frequently, loss of tone can involve only facial and neck muscles, resulting in repetitive (cluster-like) IHDs. The phenomenology of IHDs in Narcolepsy type 1 can range from episodes of subtle, slow motion and repetitive IHDs associated with the typical cataplectic face (see online supplementary video 2: segment 6a) to more abrupt and brief head nodding (see online supplementary video 2: segment 6b), or to more violent and sustained head drops (see online supplementary video 2: segment 6c).

IHDs in other conditions or in yet unclassified entities

IHDs in nodding syndrome

Nodding syndrome (NS) is an unexplained epidemic phenomenon first described in 1960 in Sudan, and later in other populations of Central Africa, with epidemic appearance of repetitive episodes of head nodding, characteristically occurring among children while eating or drinking cold water, and variably associated with other epileptic activity and neurological, cognitive and growth decline. Thousands of cases have been reported; however, only a few reported objective examinations. Polygraphic recordings disclosed epileptic abnormalities and loss of EMG tone, while neuroimaging and laboratoristic tests were inconclusive. Based on these reports, NS still represents an unexplained endemic epileptic encephalopathy.42

IHDs in Sandifer syndrome

Sandifer syndrome is an uncommon disorder mainly manifesting in infants and young children, and is characterised by abrupt onset of crying, irritability, eye/head version and intermittent abnormal postures mainly with a dystonic pattern involving the neck muscles, secondary to gastro-oesophageal reflux.43 Neck movements are due to tonic contraction of neck muscles resulting in ‘twisting’ movements of the neck. Episodes may last from a few minutes to hours and usually occur soon after a meal (see online supplementary video 2: segment 7) and may be associated with vomiting. The underlying mechanism is still unexplained. It has been suggested that the abnormal movements can be the result of a mechanism to protect the upper airways or to relieve the abdominal pain. The diagnosis is often challenging and it can frequently be misdiagnosed as paroxysmal dystonia, or epileptic or cataplectic attacks. The EEG registration could help to rule out epilepsy. Appropriate investigations to diagnose gastro-esophageal reflux disease (GERD) allow proper management and can help to avoid possible complications related to malabsorption (anaemic status, retarded growth and lung disease). (for a revision on diagnosis and practical management).44

IHDs of unknown origin in infants

IHDs may also be the expression of benign myoclonus of early infancy or of benign non-epileptic attacks in infancy (BNEAI), which are terms referring to a wide variety of episodic motor phenomena in otherwise normal babies. Episodes usually start during the first years of life and stop by early childhood. The paroxysmal episodes may indeed resemble different types of epileptic seizures, but the EEG should be clear.45

In this regard, Capovilla et al46 recently reported a novel type of BNEAI characterised by repeated head drops, mimicking epileptic negative myoclonus of the neck in his cohort. Episodes appeared in the second semester of the first year of life, were not associated with other types of spells, had no triggering factor and spontaneously disappeared within a few months. Video-polygraphic recording showed an abrupt loss of muscle tone of the neck lasting a few seconds and causing head drop. The frequency of the episodes was very high (up to hundreds daily, usually in clusters).


We show that IHDs can be associated with a wide spectrum of different disorders. Hence, there is need for a rational diagnostic work up, especially in those patients presenting with isolated IHDs. We propose a practical algorithm to guide the differential diagnosis in figure 1.

Figure 1

Diagnostic algorithm. BECTS, benign childhood epilepsy with centro-temporal spikes; BNEAI, benign non-epileptic attack in infancy; BPTI, Benign Paroxysmal Torticollis of Infancy; GERD, gastroesophageal reflux disease; MAE, Myoclonic-Astatic Epilepsy; MEI, Myoclonic Epilepsy of Infancy; PEM, Progressive Myoclonic Epilepsy; SAE, Spasm-associated encephalopaties; SOREMP, Sleep onset in REM period.

IHDs can be categorised according to the age at onset. Most of the conditions associated with IHDs occur in infancy (ie, epileptic encephalopathy, progressive myoclonic epilepsy and IHDs of unknown origin)7–17 ,45 ,46 or childhood (ie, stereotypies, paroxysmal torticollis, narcolepsy type 1 and Tourette syndrome).3–5 ,18–20 ,29–31 Conversely, the only type of IHDs occurring during adulthood is seen in choreic syndromes and sometimes in narcolepsy type 1.1 ,2 ,23–26 ,35 IHDs may, rarely, in infancy/childhood as well as in adulthood, be symptomatic of a structural brain lesion and, accordingly, as a first approach, a Brain MRI is recommended in order to exclude these forms, which frequently may by treatable by neurosurgery. In fact, on the one hand, in children who show episodes of intermittent and repetitive head drop, with a typical frequency of 2–3 Hz and without any particular trigger, a neuroimaging may show a posterior fossa lesion (bobble head syndrome).33 On the other hand, isolated IHDs occurring during adulthood are quite uncommon and neuroimaging is required to rule out symptomatic conditions, such as symptomatic epilepsy due to neoplastic or vascular lesions (especially involving the posterior fossa).

Among infancy and childhood disorders, where IHDs can represent the only or major clinical sign, it is first important to rule out an epileptic origin by means of EEG-EMG polygraphic recording plus neuroimaging investigation. In fact, epileptic disorders are, by far, the commonest conditions associated with IHDs. Once an epileptic origin is excluded, the clinician has to be aware of at least one other rare but treatable condition that can be treatable. Indeed, IHDs manifesting as wriggling and twisting movements of the neck, and occurring during meals or soon after, would be quite suggestive of Sandifer syndrome and thus laboratoristic and instrumental examinations looking for GERD should be pursued accordingly.43 ,44

More commonly, however, there are additional symptoms or signs on examination that can help in the differential diagnosis. For instance, the association with other seizures or mental impairment would point toward an epileptic origin,8–10 ,17 the occurrence in the setting of hypersomnolence or sleep attacks suggests a cataplectic origin,3 ,5 ,35 ,39 while the presence of other MDs (ie, other choreic movements, tics, etc)1 ,2 ,18 ,25 ,26 suggests a MDs syndrome (figure 1).

Regardless of the age at onset, polygraphic recordings would help in distinguishing the atonic variant (ie, epileptic myoclonic encephalopathies, benign non-epileptic attacks of infancy and cataplexy) from all the other forms, which will be, instead, tonic in nature. Additionally, IHDs’ duration can also be helpful. The briefest occur in myoclonic seizures and epileptic spasms (millis),7 ,15 ,16 the intermediate in cataplectic episodes,3 ,4 ,37 ,39 stereotypies,19 ,20 tics18 and chorea (seconds to minutes),1 ,2 ,25 ,26 and the longest are seen in paroxysmal dyskinesia,30–32 or Sandifer syndrome (minutes to hours).43 ,44 The particular phenomenology of IHDs can also aid in the differential diagnosis. Indeed, IHDs appear more ‘violent’ and sustained in choreic syndromes, especially in the contest of CA,1 and in tic disorders/stereotypies,18–20 than in the context of epilepsy7 ,9 ,10 ,15–17 or cataplexy.3 ,4 ,35 ,37 Recognition of a trigger is valuable (positive emotion in the cataplectic episodes),37 ,39 a performed movement in some types of paroxysmal dyskinesia,27 ,28 or, in a broad sense, an ‘inner urge’ in tics disorders18). Finally, except for the ‘headbanging’, which manifests during drowsiness and sleep,35 ,36 and seizures, which can occur during sleep, in all the other settings, IHDs occur during wakefulness. However, the level of vigilance can be altered during IHDs, so that epileptic IHDs can occur in the setting of impaired consciousness while cataplectic IHDs can appear either in the context of high arousal and emotions or of remarkable sleepiness.3 ,37 ,39 ,40


This review describes the clinical spectrum of IHDs. It appears that even after excluding an epileptic origin, the conditions that can be associated with IHDs are still wide. However, clinical information (age at onset and associated symptoms and signs) and neurophysiological data (background and ictal EEG activity, EMG muscle bursts duration, etc), will help distinguish these conditions. Brain MRI, however, is always mandatory in order to exclude symptomatic cases.

It was beyond the aims of this review to provide an account of the pathophysiology of such a distinctive phenomenon. However, it may be speculated that the involvement of prominent rostro-proximal muscles can result from dysfunction at multiple nodes (mainly involving the basal ganglia and diencephalon), which, in turn, can trigger or release quiescent brainstem rhythmic motor pattern (ie, the ‘mesencephalic locomotor region’) or learned-behaviours that in healthy subjects are physiologically suppressed.47 ,48


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Contributors EA was involved in the conception, organisation, execution; manuscript preparation writing of the first draft and review and critique. GP was involved in the conception, organisation, manuscript preparation, review and critique. RE was involved in the execution, manuscript preparation, review and critique. PT was involved in the organisation. BB was involved in the execution. RL was involved in the organisation. KPB was involved in the conception, organisation; manuscript preparation, review and critique.

  • Competing interests Financial disclosures: KPB has received honoraria/financial support to speak at/attend meetings from GSK, Boehringer-Ingelheim, Ipsen, Merz, Sun Pharma, Allergan, Teva Lundbeck and Orion pharmaceutical companies. KPB holds grants from MRC Wellcome Strategic grant (Ref number: WT089698) and PD UK (Ref number: G-1009). KPB and BB hold a grant from the Gossweiler Foundation. All the other authors have nothing to disclose.

  • Patient consent Obtained.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.