Statistics from Altmetric.com
Falls are frequent in the elderly, often leading to significant injury, transfer to residential care or death. Well-recognised risk factors include medications (particularly polypharmacy and the use of psychotropic agents), cognitive impairment, pre-existing gait disturbance, age-related muscular wasting (sarcopenia), dizziness and syncope (box 1). However, the cause of many falls, described by WB Matthews1 as ‘The Undiagnosable Blackout’, remains unexplained. When the fall is sudden and not associated with a perceptible loss of consciousness, it has been described as a drop attack.2
Aetiology of drop attacks
Ménière’s disease — otolithic crisis
Superior canal dehiscence syndrome
Atherosclerotic vertebrobasilar insufficiency
Craniocervical junction pathology
Chiari I malformation
Posterior fossa tumour
Osteophytes resulting in brainstem compression
Focal motor seizures
Colloid cyst of the third ventricle
Syncope with unrecognised loss of consciousness
To our knowledge, this term was introduced by Sheldon2 in 1960, in describing elderly patients who fell, often with preservation of consciousness, but who were unable to rise for minutes, or sometimes hours. Greenwood and Hopkins3 later commented that some, although remaining conscious, appeared unable to generate sufficient antigravity muscle tension in time to prevent falling. A correlation between postural instability and the tendency to trip with age, found by Overstall et al,4 did not explain these patients’ inability to avoid a fall.
Sheldon2 had commented that older people frequently reported inability, after tripping, to preserve their balance—saying ‘once you’re going, you’ve got to go’. This remark reveals a pre-existing age-related problem with balance. He noted ‘they appear to fall under the unrestrained pull of gravity and the speed of descent is such that injuries are common…there is little doubt that the loss of power is associated with a loss of muscle tone – a flaccid state’. Sheldon considered that this was due to a disturbance of function, associated with age-related changes in the cerebellum, and in the reticular substance of the brainstem.
Although many conditions can cause drop attacks (box 1), we postulate that myoclonus might be an under-recognised cause or contributor.
Having had four such falls at an age appropriate for the subject of this article, JWL noted that he had experienced a sequence of events which closely resembled the jerks and falls of patients suffering from familial myoclonic epilepsy5 or from posthypoxic encephalopathy.6 To investigate a possible epileptic mechanism for falling attacks in the elderly, he enlisted the aid of a younger colleague (SEW) in assembling and comparing relevant publications. Although falling attacks had formerly been considered a possible manifestation of epilepsy, the notion was discarded because of the poor response of drop attacks to standard anticonvulsants. Because of recent reports of myoclonus responding to piracetam7 and levetiracetam,8 this approach now warrants review.
Some years ago, walking briskly up a cobblestone road in the Cinque Terre, I stumbled and fell, with no warning. I was immediately surrounded by sympathetic pedestrians blaming the uneven surface. As I had no recollection of falling, I was not convinced. I was soon able to stand and to walk, at a more sedate pace.
The second episode occurred some years later, by which time my balance had deteriorated such that I required the use of a walking stick. While walking on a flat carpet, my stick struck a minor obstruction, making me startle, lurch to the right and make a series of short, lateral, jerking steps before my legs gave way and I fell. I can remember falling while attempting successfully to rotate my trunk to strike the floor with my right shoulder rather than my head. Although my legs were weak, I dragged myself to a chair on which I rested my arms until leg power returned and I could stand.
The third episode occurred when I unexpectedly slipped on a bathroom floor and fell into a sitting position. I was unable to stand. I rotated my body to propel myself to the edge of the bath by sliding over the slippery floor. After I rested my arms on the side of the bath for several minutes, power returned to my lower limbs and I could stand.
On two occasions in the past year, I have fallen after slipping on a smooth concrete floor. A startle response was followed by lurching to the right, without the ability to contract my antigravity muscles. I fell heavily, abrading my right upper arm. On each of these two occasions, I remember falling. After the first fall, I could get to my feet immediately. After the second, I was unable, for a few minutes, to move. My upper limb muscles also felt weak, but my lower limb muscles felt paralysed—so that I had dropped like a stone. I then rolled into the prone position, pushed myself up as strength returned and then scrambled to my feet, albeit shakily.
Physiology of maintaining posture
Standing upright requires continuous, repetitive (‘tonic’) discharge of motor neurons supplying the antigravity muscles in the trunk and lower limbs.
The term ‘muscle tone’ is often used clinically to describe the sensation of firmness felt on palpation of a muscle belly while its fibres are contracting, . Muscle tone is mediated by sustained activity in the stretch reflex arc by stretch (or vibration) of sensory receptors in the muscle spindles, tension of which is maintained by their gamma efferent innervation (alpha-gamma linkage). PBC Matthews9 demonstrated that the application of vibration to the soleus muscle of the decerebrate cat caused a sustained contraction, later termed a tonic vibration reflex (TVR), of that muscle. This suggested that studying the TVR of the antigravity muscles in a suitable animal model could help in elucidating the mechanism of supraspinal control of muscle tone in man.
To examine the part played by individual supraspinal structures in anaesthetised cats, muscle contraction in an antigravity muscle was assessed, while brainstem nuclei and tracts were stimulated stereotactically, by measuring the tension exerted by the tendon of the muscle active during a TVR.11 12
Muscle tone was found to be augmented by stimulation of the lower pontine reticular formation, lateral vestibular nucleus and vestibulospinal tract, whereas it was diminished or abolished by stimulation of the medial part of the medullary reticular formation and the downstream fibres of the tract descending from it.12 These inhibitory reticulospinal fibres lie in the medial part of the internal capsule and cerebral peduncle and, separated from the fibres of the pyramidal tract, in an area medial and dorsal to the peduncle. Moreover, these inhibitory extrapyramidal fibres have been shown, in our laboratory studies,12 to be activated by stimulation of the contralateral pericruciate (sensorimotor) area of the cat cortex.
This series of experiments11 12 has thus established an animal model in which excitation of the sensorimotor cortex by startle or by the sudden execution of a voluntary movement could trigger cortical reflex myoclonus,13 which then inhibited tone in the antigravity muscles, causing an akinetic (drop) attack.
On the other hand, vestibular afferents, which project to the lateral vestibular nucleus and the vestibulospinal tract, normally maintain a tonic excitatory influence on the extensor muscles of the trunk and lower limbs, reflexively assisting the subject to resist gravity.
Aetiology of drop attacks
In a review of 93 patients with recurrent idiopathic drop attacks investigated for an underlying aetiology, Parry and Kenny14 found an attributable diagnosis in 84 subjects, with a likely cardiovascular cause in more than half (n=49). Other aetiologies included vestibular pathology, cerebrovascular disease, cerebellar degeneration, vertebrobasilar insufficiency, cervical spondylosis, visual impairment and medication-related. Dey et al 15 reported that 24 of 35 patients with ‘idiopathic drop attacks’, when investigated for an underlying aetiology, had cardiovascular syncope, including vasovagal syncope, carotid sinus syndrome or orthostatic hypotension. Thus, although patients suffering drop attacks typically report preservation of consciousness, syncope might be responsible for some episodes. A list of potential causes of drop attacks is summarised in box 1.
Drop attacks in Ménière’s disease, ‘Tumarkin otolithic crises’, were described by Tumarkin16 in 1936. These are characterised by sudden falling, without loss of consciousness, with patients frequently describing a sensation of being thrown to the ground. In Tumarkin’s original report, he described three patients experiencing sudden collapse with preservation of consciousness—the attacks coming ‘like a bolt from the blue’. It is thought that otolith stimulation plays a key role in the pathogenesis of such attacks, resulting in inappropriate activation of the vestibulospinal pathways—leading to a fall. Drop attacks might also be a feature of other forms of vestibular disease, including the superior canal dehiscence syndrome and as a sequel to transtympanic gentamicin administration.17
Vertebrobasilar insufficiency might result in impairment of posture due to transient ischaemia of the corticospinal tracts or of the reticular formation and reticulospinal tracts as a result of localised vascular disease.18 In these cases, a drop attack is rarely an isolated feature; dizziness, vertigo, headaches, vomiting, diplopia and ataxia are commonly associated.19 Transient compression of the brainstem is a well-described phenomenon which can cause a drop attack, as can be seen in patients with Chiari I malformation.
Cataplexy, specific to narcolepsy, is the sudden loss of muscle tone, often triggered by a strong, usually explosive emotion, such as laughter, excitement or anger.20 The weakness, lasting minutes or less, characteristically involves all striated muscles except the diaphragm, and is associated with preservation of consciousness.21 Thought to represent rapid eye movement (REM) sleep atonia occurring during wakefulness, it is due to inhibition of the alpha motor neurons. H reflexes are suppressed, and deep tendon reflexes are absent.
Epileptic drop attacks
Atonic seizures (epileptic drop attacks) result from sudden loss of tone in the postural muscles, sometimes preceded by a brief myoclonic jerk. These have been described in temporal, frontal and multifocal epilepsy.22
Tripping or another unexpected sudden event can, in a susceptible elderly person, induce a startle response, comprising a myoclonic jerk followed by an akinetic period—due to brainstem inhibition of the tonic contraction of the antigravity muscles—causing a drop attack (cortical reflex myoclonus).13 Subsequent muscle weakness can persist for minutes or hours, delaying rising. This resembles the sequence of events in epileptic drop attacks. Glass et al 23 described orthostatic myoclonus (myoclonus occurring only when the subject was in the upright posture). In a series of 15 patients with orthostatic myoclonus demonstrated on electromyography (EMG), 14 were aged over 65. Although the clinical presentation of most patients in this series was progressive gait decline, 5 of the 15 suffered recurrent, unpredictable falls, presumed to be related to myoclonus. The EMG in these cases was normal at rest, but demonstrated myoclonic bursts when the patients were upright.
Cryptogenic drop attacks
Despite the numerous recognised causes of drop attacks, a substantial number remain unexplained (cryptogenic). Hoeritzauer et al 24 have recently suggested that the aetiology of such cryptogenic attacks might, in many cases, be functional. They identified that 43% of patients in their cohort had prodromal dissociation or panic, and most (90%) had either comorbid functional somatic disorder, functional neurological symptom disorder or both. They highlight several features supporting their hypothesis that the symptoms have a functional basis, including the inability to recall the fall and the use, in some cases, of distraction techniques to prevent falling. In their paper, 89% of patients were female and the mean age was 44 years old. Although a functional basis might contribute to some of the recurrent attacks occurring in a younger patient population, it is unlikely that this mechanism will apply to the drop attacks we describe in the elderly.
EMG recordings during attacks
Because falling attacks of the elderly are sporadic and do not recur with a frequency to permit prior placements of electrodes, EMG recordings have been made, during similar akinetic periods, of patients with epileptic myoclonus5 or with posthypoxic myoclonus.6 In these cases, a silent period occurred abruptly in the EMG of the affected muscles, so that the limbs became atonic for up to 340 ms, causing the affected limb to drop under the influence of gravity, or the patient to fall like a dead weight. When the H reflex (roughly the electrically induced equivalent of the ankle jerk) was recorded during an akinetic attack, it persisted at lower amplitude, indicating that the responsible anterior horn cells, although deprived of their supraspinal drive, remained accessible to reflex stimulation at the spinal level.
A possible treatment
JWL was struck by the similarity between his personal experiences and the descriptions related to him by patients suffering from progressive myoclonic epilepsy5 or from posthypoxic myoclonus.6 The following were features in common:
Persistent underlying impairment of postural stability, age-dependent in the case of falling attacks, and, in epileptic myoclonus, sometimes accompanied by frank cerebellar signs.
Precipitation of the fall by a quick voluntary movement, or by a sudden startle such as a walking stick touching an unexpected impediment, or a by threatened fall on a slippery surface.
Brief contraction of the antigravity muscles, followed immediately by loss of power in the lower limb muscles, so that no stabilising movement was possible, leaving the subject to fall heavily to the ground (cortical reflex myoclonus).
Piracetam has proven efficacy in patients with myoclonic epilepsy,7 and, in our clinical experience (unpublished work), was also shown to be partly effective in the management of posthypoxic myoclonus. Although there are, as yet, no prospective trials evaluating the efficacy of any treatment for this condition, the related agent, levetiracetam, has similarly been used, with anecdotal success, in patients with orthostatic myoclonus.8 Given the favourable side-effect profile of both piracetam and levetiracetam, and the clinical similarities between our patients’ reports and the manifestations and patients’ experiences of other forms of myoclonus, elderly patients suffering frequent drop attacks might respond favourably to a trial of these medications.
Although there are many causes of falls in the elderly, we propose that myoclonus might play a part in the various mechanisms. Greater recognition of its influence could lead to more appropriate, targeted treatment. As patients with various forms of epileptic myoclonus have shown some response to levetiracetam and similar agents, a clinical trial of regular medication in elderly patients subject to frequent unexplained falling attacks would be worth undertaking.
We are grateful to Peter C Arnold, OAM, of Sydney, for editorial support in preparing this paper.
Contributors JWL is the primary author, and contributed to the introduction, clinical observations, and all references to EMG recordings. SEW is the secondary author, and also contributed to the introduction and clinical observations, as well as the treatment review of myoclonus and the conclusion.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
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.