The electrophysiologic evaluation of movement disorders is not used as widely as traditional electromyography (EMG) and nerve conduction studies, but it can contribute to clinical management. Physiologic analysis can be useful in the classification of difficult involuntary movements by providing information that is impossible to obtain by clinical observation alone. Physiologic studies can also be valuable for guiding therapy and for increasing our understanding of the pathophysiology of movement disorders. Additionally, physiologic analysis permits the quantification of involuntary movements, which is particularly valuable for research purposes such as in drug trials.
Analysis of movement
Measurement of Movement
A direct measurement of movement can be made by recording the angular changes at joints. An electrical signal proportional to the angular position of a joint can be produced, for example, by a device incorporating a potentiometer with its axis aligned to the center of the joint. Velocity and acceleration can be obtained by differentiation. Movement also can be measured with video systems; these are used commonly for complex movement such as gait. Movement of the hand in two dimensions can be assessed with a data tablet. Acceleration can be determined directly with an accelerometer. Accelerometers in common use measure movement only in one axis, so they must be oriented carefully in the direction of movement. Also available are triaxial accelerometers that measure movement in three orthogonal axes and permit three-dimensional description of movement. Accelerometers are used most commonly for the analysis of tremor. Less commonly used are gyroscopic devices that measure rotations. There are also triaxial gyroscopes, and even devices with both triaxial accelerometers and triaxial gyroscopes that then permit a complete description of translations and rotations.
The output of an accelerometer is a time-varying analog signal proportional to instantaneous acceleration ( Fig. 20-1 ). In the case of tremor, the signal can be processed in several ways to get a number (or numbers) to characterize the waveform. The root-mean-square (RMS) amplitude calculated over a period (e.g., 20 to 30 seconds) is the best measure of the “average” amplitude. A more complete analysis of the waveform can be made by calculating the Fourier transform (usually accomplished by computer using a fast Fourier transformation, or FFT). This produces a plot of power in the signal as a function of frequency. If the signal has one major frequency component, the plot will have a single major peak; the power of that peak will be proportional to the RMS amplitude. The acceleration signal can be double-integrated to produce the total distance moved; if only the portion of the signal in the tremor peak is processed in this way, the result is the distance moved by virtue of tremor.
Another measure of movement is the electromyogram of the muscles producing the movement. Because EMG is a direct measure of alpha motor neuron activity, it provides information about the central nervous system command that generates the movement. Numerous muscles act on each joint; it is necessary to record from at least two muscles with antagonist actions. EMG is used occasionally as a measure of force, even though the relationship between EMG amplitude and force is only approximate. The timing information from the EMG signal is much more accurate.
EMG data can be measured with surface, needle, or wire electrodes. The advantages of surface electrodes are that they are not painful and record from a relatively large volume of muscle. The advantage of needle electrodes is that they are more selective; this is sometimes a necessity when recording from small or deep muscles. Traditional needle electrodes are stiff, and it is best to use them when recording from muscles during movements that are close to isometric. Pairs of fine wire electrodes have the advantage of selectivity similar to that of needle electrodes and are also flexible, permitting free movement with only minimal pain. In any case, it is important to avoid movement artifact, which can contaminate the EMG signal. Wire movement should be limited. Low-frequency content of the EMG signal can be restricted with filtering. Impedance of surface electrodes should be reduced.
Inspection of the EMG signal of an involuntary movement reveals, first, whether the movement is regular (usually a tremor) or irregular. There are sometimes surprises in such an analysis. Rhythmic EMG activity can appear irregular clinically if the amplitude varies; irregular EMG activity sometimes will appear rhythmic clinically if it is rapid. The duration of the EMG burst associated with an involuntary movement can also be measured; specific ranges of duration are associated with different types of movements. Specification of duration in the range of 30 to 300 msec merely by clinical inspection is virtually impossible because of the relative slowness of the mechanical events compared with the electrical events. Finally, antagonist muscle relationships can be specified as synchronous or asynchronous (reciprocal) by inspection of the EMG signal. In a tremor, asynchronous activity would be described as alternating.
Assessment of Voluntary Movement
As with involuntary movements, it is possible to measure voluntary joint movement directly. Movement can be analyzed at a single joint or at several joints at the same time. Measurements can be made of the movement time (i.e., time from initiation to completion of a movement) and accuracy. If the movement is triggered by a stimulus, then the reaction time (i.e., time from stimulus to initiation of movement) also can be measured.
EMG patterns with voluntary movement also provide valuable information. Normal patterns vary with the speed of movement ( Fig. 20-2 ). A slow, smooth movement is characterized chiefly by continuous activity in the agonist. A movement made as rapidly as possible (a so-called ballistic movement) has a triphasic pattern with a burst of activity in the agonist lasting 50 to 100 msec; a burst of activity in the antagonist lasting 50 to 100 msec; and return of activity in the agonist, often in the form of a burst.
Tremors can be divided into two types: those at rest and those seen with action. “Rest” is only relative; some slight tonic postural maintenance often is required. Action tremors must be subdivided into those seen just with postural maintenance (postural or static tremor) and those requiring goal-directed movement (intentional or kinetic tremor). A third division of action tremors is those seen only with specific types of kinetic movement, such as handwriting, and is called task-specific tremor.
The physiology differs in different forms of human tremor. Tremors may come from mechanical oscillations, mechanical-reflex oscillations (EMG activity is entrained with a mechanical oscillation), normal central oscillators, and pathologic central oscillators ( Fig. 20-3 ). An excellent method for determining the physiology consists of combined accelerometry and EMG using spectral analysis and study of the tremor with and without weighting of the body part. This allows separation of tremors of mechanical and mechanical-reflex origin from those brought about by central oscillators because mechanical and mechanical-reflex tremors have a slower frequency with limb weighting.
Parkinsonian Tremor at Rest
The most common tremor at rest is that seen with Parkinson disease or other parkinsonian states, such as that produced by neuroleptics. The tremor usually is seen in the context of other basal ganglia symptoms, but on rare occasions it can be the sole clinical finding. It is present at rest and disappears with action, but it may resume with a static posture, particularly late in the disease. The frequency is 3 to 7 Hz. EMG studies show antagonist muscles to be active alternately ( Fig. 20-4 ). The tremor frequency is not altered by weighting and hence has its origin in a central oscillator, but its location is not known. Postural tremors may occur in Parkinson disease, as well as a rest tremor.
Exaggerated Physiologic Tremor
Physiologic tremor is a normal postural action tremor. In certain circumstances (e.g., anxiety, fatigue, thyrotoxicosis, and excessive use of caffeine), the tremor can be increased in magnitude and may be symptomatic. The frequency usually is in a range from 5 to 12 Hz, varying in part because of the weight of the tremulous body part. The EMG in mild cases may look just like a normal interference pattern without well-defined bursting. In more severe cases, bursting may appear; it is usually synchronous in antagonist muscles. The accelerometric and EMG spectral peaks will be the same and will shift together with weighting. There may also be a component from an 8- to 12-Hz central oscillator in this condition.
Essential tremor is a common neurologic disorder that often runs as an autosomal-dominant trait in families. It may appear in childhood or late in life and runs a slowly progressive course. Typically, it is a postural tremor; in some patients there is some increase in tremor with intention (kinetic movement), and in others the tremor occurs primarily with goal-directed movement (intentional essential tremor). Rarely, it appears to persist with rest. In most circumstances it is seen as the sole neurologic abnormality; there are pathologic changes in the cerebellum in many patients, but the pathophysiology is not clear. The frequency ranges from 4 to 12 Hz. EMG studies commonly show synchronous activity in antagonist muscles ( Fig. 20-5 ), but alternating activity is also possible. Sometimes it is clinically difficult to separate exaggerated physiologic tremor from essential tremor. Using accelerometry and EMG, there is a constant frequency with weighting, which is usually different from that seen with exaggerated physiologic tremor. This observation indicates that essential tremor originates from a generator in the central nervous system. The cerebellum and cerebellar circuits seem involved, and one commonly considered candidate is the inferior olivary nucleus.
Tremor with cerebellar lesions can be postural, as well as the better-known kinetic tremor. Postural cerebellar tremor can be separated into two groups designated as mild and severe. The more characteristic is severe postural cerebellar tremor, which may be present also at rest, persists or worsens with goal-directed movement, and is associated with dysmetria. It has been called rubral tremor, an inaccurate term because the responsible lesion is often in the superior cerebellar peduncle. More frequently, it is referred to as Holmes tremor. Typically, it has a frequency of 2.5 to 4 Hz, affects proximal muscles more than distal muscles, waxes and wanes, and has a tendency to increase progressively in amplitude with prolonged posture. EMG studies show bursts of activity lasting 125 to 250 msec and alternation of activity in antagonist muscles. The most common etiology in general neurologic practice may be multiple sclerosis, but there are other causes such as stroke, head trauma, or tumors.
Mild postural cerebellar tremor is less well defined. The group includes tremors that are transient and more rapid (up to 10 Hz), and that have distal predominance.
Kinetic tremor without postural tremor usually is ascribed to cerebellar dysfunction and is often called cerebellar intention tremor. It may occur also with a postural tremor. The lesions can be in the cerebellum or cerebellar pathways. Kinetic tremor is characterized by rhythmic oscillations about the target of movement; EMG studies show alternating activity in antagonist muscles. It should be differentiated from sequential irregular, inaccurate movements toward the target, which have been named serial dysmetria ( Fig. 20-6 ).
Wilson disease can present with tremor as its sole manifestation, although other movement disorders and psychiatric disturbances are commonly present as well. The tremor is an action tremor present with posture and kinetic movement. Physiologic analysis shows alternating activity in antagonist muscles at 3 to 5 Hz.
Postural tremor may be seen in the setting of congenital or acquired peripheral neuropathies. The pathophysiology is obscure. The tremor seen with hereditary sensorimotor neuropathy type 1 simply may be essential tremor that is co-inherited. Physiologically, the frequency is in the range of 6 to 8 Hz, and EMG studies show a mixture of synchronous and alternating activity in antagonist muscles. It is likely that, in many circumstances, slowing of nerve conduction in the peripheral loop will give rise to delays that create instability. In other circumstances, studies suggest a coexisting central oscillator.
Palatal tremor, also known as palatal myoclonus, is characterized by rhythmic movements of the palate at approximately 3 Hz. There are two separate disorders: essential palatal tremor, which manifests an ear click; and symptomatic palatal tremor, which is associated with cerebellar disturbances. The palatal movements are caused by activation of the tensor veli palatini muscle in essential palatal tremor and of the levator veli palatini muscle in the symptomatic disorder. In symptomatic palatal tremor, the palatal movements may be accompanied by synchronous movements of adjacent muscles such as the external ocular muscles, tongue, larynx, face, neck, diaphragm, or even limb muscles. Symptomatic palatal tremor is associated with hypertrophy of the inferior olive and many authorities consider it to arise there, but the generator for essential palatal tremor probably differs.
Primary writing tremor is task-specific and appears only with handwriting and a few additional skilled tasks. It is not seen with all skilled tasks and is not produced by posture or goal-directed movement in general. The condition probably is underdiagnosed because it is often confused with essential tremor and tremulous writer’s cramp. Although the originally described patient had tremor with synchronous activity, subsequent experience suggests that the tremor is more commonly alternating in antagonist muscles with a frequency of 5 to 6 Hz.
Orthostatic tremor is a tremor of the legs that occurs only when standing; it is not present when voluntary movements of the legs are made while lying down, nor is it present with walking. The tremor in leg muscles is at about 16 Hz, is not influenced by peripheral feedback, and is synchronous between homologous leg muscles. The site of the central generator is unknown.
Tremor can be a conversion symptom. Such tremors can take many forms, but the most common are action tremors with alternating activity in antagonist muscles. Often they violate rules of clinical behavior with sudden starts and stops, and disappearance with distraction, but the distinction can be difficult. Psychogenic tremors vary in amplitude more than expected. Clinical neurophysiologic assessment can be very helpful. Accelerometry in most tremors shows a narrow peak frequency that does not change over short periods; in psychogenic tremors the peak may be broad and change frequently. While organic tremors differ in frequency in different limbs, psychogenic tremors are more commonly exactly the same frequency. An important feature useful for physiologic analysis is that the tremor frequency may be entrained or altered by requested tapping at different rates by another body part.
Other involuntary movements
Three EMG patterns may underlie involuntary movements. One pattern, which can be called tonic , resembles slow voluntary movements and is characterized by continuous or almost continuous EMG activity lasting for the duration of the movement, from 200 to 1000 msec or longer. Activity can be solely in the agonist muscle, or there can be some co-contraction of the antagonist muscle with the agonist. Another pattern, which can be called ballistic , resembles voluntary ballistic movements with a triphasic pattern. There is a burst of activity in the agonist muscle lasting 50 to 100 msec; a burst of activity in the antagonist muscle lasting 50 to 100 msec; and then return of activity in the agonist, often in the form of another burst. The third pattern, which can be called reflex , resembles the burst occurring in many reflexes, including H reflexes and stretch reflexes. The EMG burst duration is 10 to 30 msec, and EMG activity in the antagonist muscle is virtually always synchronous.
Myoclonus is characterized by quick muscle jerks, either irregular or rhythmic. There are many types of myoclonus and no common etiologic, physiologic, or therapeutic features that bind them together. Myoclonus can be focal, involving only a few adjacent muscles; generalized, involving many or most of the muscles in the body; or multifocal, involving many muscles but in different jerks. Myoclonus can be spontaneous; activated or accentuated by voluntary movement (action myoclonus); and activated or accentuated by sensory stimulation (reflex myoclonus). Rhythmic (segmental) myoclonus has the appearance of a rest tremor but typically is unaffected by action, stimulation, or even sleep. In this disorder, a segment of the spinal cord (spinal myoclonus) or brainstem (palatal myoclonus) produces persistent rhythmic repetitive discharges usually unaffected by sleep. A number of contiguous muscles produce synchronous contractions at a rate of 0.5 to 3 Hz. Because of the slow speed of the movements, palatal myoclonus is often called palatal tremor.
By defining epileptic myoclonus as myoclonus that is a fragment of epilepsy, it is possible to divide irregular myoclonus into epileptic and nonepileptic myoclonus. The physiologic characteristics of epileptic myoclonus are as follows: (1) EMG burst length of 10 to 50 msec; (2) synchronous antagonist activity; and (3) an electroencephalographic (EEG) correlate. (The technique of EMG–EEG correlation is described later.) The EMG shows a reflex pattern. Nonepileptic myoclonus shows the following: (1) EMG burst lengths of 50 to 300 msec; (2) synchronous or asynchronous antagonist activity; and (3) no EEG correlate ( Fig. 20-7 ). The EMG patterns are either ballistic or tonic.
Examples of epileptic myoclonus are cortical reflex myoclonus, reticular reflex myoclonus, and primary generalized epileptic myoclonus; these are discussed later in relation to EMG–EEG analysis. Examples of nonepileptic myoclonus include dystonic myoclonus; essential myoclonus (e.g., ballistic movement overflow myoclonus); exaggerated startle; physiologic phenomena (e.g., hypnic jerks); and periodic movements of sleep. Frequent myoclonus may have the appearance of tremor. In the case of action myoclonus, this may be confusing clinically, but EMG analysis is definitive ( Fig. 20-8 ). Psychogenic myoclonus is a common presentation of psychogenic movement disorders, and the burst pattern is always nonepileptic in type.
Tics are quick, involuntary, repetitive movements that occur at irregular intervals. The unique feature of a tic is that it is not completely involuntary. Most patients describe a psychic tension that builds up inside them and can be relieved by the tic movement. Hence the tics can be suppressed voluntarily for some time at the expense of increasing psychic tension; patients “let the tic happen” (or perhaps even “make the tic”) to relieve the tension. Tic movements, which can be simple or complex, look like quick voluntary movements both clinically and electromyographically. EMG bursts vary from 50 to 200 msec in duration and may have a ballistic or tonic pattern.
Dystonia and Athetosis
The involuntary movements of dystonia and athetosis are similar, and the use of one term rather than the other often seems more a matter of situation and semantics than of physiology. The movements are typically slow but can be quick and may be sustained for a second or longer when the involuntary contraction is at its maximum. Dystonia often contains some sustained postures, while athetosis is slow, continuous movement. Dystonia often is used to describe proximal twisting movements and athetosis for more distal “flowing” movements. Dystonic and athetotic movements often are characterized by co-contraction of antagonist muscles. Although normal voluntary movement commonly is characterized by reciprocal inhibition, there may be some co-contraction. The co-contraction of dystonia and athetosis is excessive, with the appearance of increased tension at the joint. Some dystonic and athetotic movements are fully involuntary, arising at rest independent of will. Other movements arise as excessive, unwanted concomitants to voluntary movements. This phenomenon is called overflow, with the implication that the motor control command is sent to too many muscles with too much intensity. All of this may be caused by a failure of surround inhibition in the motor system. EMG studies can document these phenomena. The shortest EMG bursts seen even with dystonic myoclonus are in the range of 100 to 300 msec.
Dyskinesia, Ballism, and Chorea
Dyskinesia describes choreic movements seen in selected circumstances (e.g., a late consequence of neuroleptic drugs or with levodopa toxicity). Ballism describes wild, large-amplitude choreic movements; these usually involve one side of the body and then are called hemiballismus. The most appropriate adjective to describe chorea is “random.” Random muscles throughout the body are affected at random times and make movements of random duration. Movements can be brief (e.g., myoclonus) or long (e.g., dystonia). Usually, they are totally beyond voluntary control, but in some mild cases the movements can be suppressed temporarily. EMG patterns are reflex, ballistic, and tonic ( Table 20-1 , Fig. 20-9 ).