+Model
ARTICLE IN PRESS
NEUCLI-2402; No. of Pages 8
Neurophysiologie Clinique/Clinical Neurophysiology (2014) xxx, xxx—xxx
Disponible en ligne sur
www.sciencedirect.com
REVIEW / MISE AU POINT
Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus Neurophysiologie clinique des mouvements anormaux psychogènes : tremblement et myoclonies E. Apartis Inserm-UPMC UMRS 975-CRICM, Department of Physiology, Saint-Antoine Hospital, AP—HP, 184, rue du Faubourg Saint-Antoine, 75012 Paris, France Received 17 July 2013; accepted 25 August 2013
KEYWORDS Movement disorders; Psychogenic; Tremor; Myoclonus; Reflex; Movement preparation; Polymyography; Frequency
Summary Tremor and myoclonus are very common manifestations of psychogenic movement disorders (PMD). In this context, recording of movement disorders aims to provide objective criteria for a positive diagnosis of PMD, independently of the psychological situation. Neurophysiological observations are therefore considered to have a huge impact both on diagnosis and on therapeutic approaches. A specific recording strategy should be employed whenever the medical history or clinical clues raise the eventuality of a PMD. Polymyography coupled to accelerometry is used to demonstrate the major electrophysiological criteria of psychogenic tremor, namely spontaneous variability of tremor frequency and frequency entrainment induced by contralateral rhythmic tasks. Other features, such as paradoxical increase of tremor amplitude with mass loading, co-activation preceding tremor onset and alteration of voluntary contralateral motor performances when tremor is present, are also of interest. The clinical presentation of psychogenic myoclonus is extremely rich and polymorphous and can mimic virtually all forms of cortical, subcortical or spinal myoclonus. Focal, multifocal, axial or generalized jerks can occur. Psychogenic jerks can be sporadic or repetitive, rhythmic or arrhythmic, spontaneous or stimulus-induced. All of these parameters are crucial to determine an individualized neurophysiological strategy. Polymyography is critical to identify a ballistic pattern or a discordant or non-reproducible temporo-spatial organisation of the jerks, but has usually to be completed by other potentially decisive approaches. Reflex psychogenic myoclonus for example displays stimulus-response delays that are too long and variable. Spontaneous psychogenic jerks may be also preceded by a pre-movement potential, detectable by jerk-locked-back-averaging methods. © 2013 Elsevier Masson SAS. All rights reserved.
E-mail address: [email protected] 0987-7053/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.neucli.2013.08.014
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
2
E. Apartis
MOTS CLÉS Mouvements anormaux ; Psychogène ; Tremblement ; Myoclonies ; Préparation du mouvement ; Polymyographie ; Fréquence ; Réflexe
Résumé Les mouvements anormaux psychogènes prennent très fréquemment la forme de tremblements ou de myoclonies. L’objectif de l’enregistrement des mouvements anormaux est d’apporter des arguments objectifs de diagnostic positif, indépendamment du contexte psychologique. Les observations neurophysiologiques peuvent avoir un impact important sur l’issue diagnostique et la démarche thérapeutique. Une stratégie spécifique d’examen doit être mise en œuvre dès que la nature psychogène du mouvement est évoquée sur des indices cliniques. La polygraphie musculaire couplée à l’accélérométrie est l’examen de base qui permet de démasquer les deux critères électrophysiologiques majeurs de tremblement psychogène, que sont la variabilité spontanée de fréquence et l’entraînement de fréquence induit par des tâches motrices compétitives rythmiques, ainsi que l’augmentation d’amplitude sous l’effet du poids, le signe de co-activation et l’altération des performances motrices controlatérales en présence du tremblement. La présentation clinique des myoclonies psychogènes est riche et polymorphe et peut mimer différentes formes de myoclonies, corticales, sous-corticales ou spinales. Les myoclonies psychogènes peuvent être focales, multifocales, axiales ou généralisées ; leur survenue peut être sporadique ou répétitive, rythmique ou arythmique, spontanée ou induite par les stimulations. Tous ces paramètres sont essentiels pour déterminer une stratégie neurophysiologique individualisée. La polymyographie est essentielle pour identifier un pattern balistique ou une organisation temporo-spatiale discordante ou non reproductible, mais doit être généralement complétée par d’autres approches. Les myoclonies psychogènes réflexes sont caractérisées par des délais de réponse aux stimulations longs et variables. La présence d’un potentiel de préparation au mouvement apporte un argument en faveur de la nature psychogène de myoclonies spontanées. © 2013 Elsevier Masson SAS. Tous droits réservés.
Introduction Psychogenic movement disorders (PMD) are common. As highlighted by Brown and Thompson [4], fifty percent of patients with a PMD improve to some degree and only onethird resolve. These patients with better outcome tend to have a shorter duration of movement disorder, suggesting that early diagnosis and treatment may be important. Clinical features suggestive — but not diagnostic — of PMD are the sudden onset of an inconsistent and variable movement disorder, which is incongruous with typical ‘‘organic’’ movement disorder. Other clues are marked reduction of the abnormal movement with distraction, dramatic increase in severity and complexity of movement induced by direct observation and the occurrence of spontaneous periods of remission. The evidence of underlying psychopathology and associated psychogenic symptomatology is also suggestive, but not necessarily diagnostic of a psychogenic etiology. More convincing, if detected, is the disappearance of the movement disorder when supposedly unobserved or following suggestion or placebo [4,31]. The neurophysiological examination is regularly crucial in diagnosing PMD, assisting the clinician in evaluating these patients. In this setting, the recording of movement disorders aims to provide objective criteria for a positive diagnosis of PMD, by using measurable features, independently of the psychological context. Neurophysiological observations thus have a huge impact on both diagnosis and therapeutic approaches. Additionally, in some cases, visualisation of neurophysiological data helps the physician guide the patient towards better understanding of his symptoms, opening a window towards recovery. A specific recording strategy should be performed whenever the medical history or clinical clues raise the possibility of a PMD or when the electrophysiological criteria of an
organic abnormal movement cannot be identified. As clinical neurophysiological testing can be helpful mainly in diagnosing psychogenic myoclonus and tremor, and can lead to definite diagnosis, we will focus on both these aspects in this review.
Psychogenic tremor Tremor is a frequent manifestation of PMD. After a reminder of the clinical clues of psychogenic tremor, the neurophysiological tools that should be used to reveal objective positive signs in comparison to the classical characteristics of tremors of organic origin will be detailed in the following section. Psychogenic tremor should be distinguished chiefly from pyramidal clonus and organic tremors such as enhanced physiologic tremor, parkinsonian tremor, dystonic tremor, and odd tremors, in order to allow optimal management. Aside general clinical features, relevant criteria revealed by electrophysiological study in psychogenic tremor are: • spontaneous variability of tremor frequency; • frequency driving induced by contralateral rhythmic tasks; • paradoxical increase of tremor frequency and amplitude with mass loading, and; • co-contraction preceding tremor onset (co-activation sign). By contrast, organic tremors have a stable frequency that cannot be driven to another one and are classically amplified by contralateral tasks. When clinical red flags are present, a specific recording protocol should be proposed in order to unmask these features. This is based on simultaneous
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders recording of the limb concerned by the abnormal movement with an accelerometer (Acc) and surface electromyography (EMG) of a few relevant muscles, which allows both measurement of tremor frequency and EMG burst duration, and study of the pattern of contraction between antagonist muscles. Simultaneous recording of the contralateral limb is critical to demonstrate the frequency driving induced by a variety of manoeuvres such as finger tapping or other repetitive movements in response to an external signal from a metronome. Methods of signal analysis, including timefrequency analysis of EMG and inter-limb EMG coherency are helpful to answer these questions. The former allows better definition of the frequency content of the tremor while the latter demonstrates the temporal link between two distant rhythmic activities. When a high coherency level evidences such a temporal link in two different limbs, one can hypothesize that both rhythmic activities have a common generator in the brain or depend on two coupled systems. For example, when a control subject voluntarily drums to the beat at the same frequency with both hands, right-left coherency is high. By contrast, there is no right-left coherency in essential or parkinsonian tremors, since in these situations two different central uncoupled oscillators are involved for different tremulous limbs [22]. Let us now describe more precisely what are classically considered as neurophysiological signs leading to a positive diagnosis of psychogenic tremor.
Tremor frequency and its spontaneous variability The frequency of psychogenic tremor is comprised between 4 and 10 Hz [5], and there is a large overlap between psychogenic tremor and organic tremor frequencies, except for high-frequency (14—18 Hz) orthostatic tremor. Low frequency (3 Hz) psychogenic tremors are rare, and this frequency is more evocative of mesencephalic or cerebellar tremors. The upper frequency limit of voluntary mimicked tremor is 11 Hz, so that any tremor that is faster than 12 Hz is very likely to have an organic aetiology [4]. As it is very difficult to maintain a voluntary oscillatory limb movement at the same frequency for any length of time, frequency variability in the same limb segment and in the same postural condition is strongly suggestive of a psychogenic tremor. Zeuner et al. [33] proposed 1.5 Hz as the lower limit of this frequency variability in psychogenic tremor. When variability is not spontaneously present, it can be elicited by mental or contralateral motor manoeuvres, the aim of which is to distract the patient. These manoeuvres should preferably be complex, in order to sufficiently engage attentional resources.
Influence of contralateral rhythmic tasks on tremor and inter-limb frequency dissociation An important sign of psychogenic tremor relates to the difficulty in voluntarily maintaining two or more unrelated, non-harmonic, rhythms in different body parts [21]. Most people seem incapable of producing more than one bimanual frequency. Only trained musicians are able to maintain bimanual polyrhythms. Therefore, the simultaneous occurrence in separate limbs of tremors with different
3 frequencies (frequency dissociation) is highly suggestive of organic tremor [4]. Assuming that certain motor systems are disrupted in Parkinson’s disease and essential tremor, it is proposed, on one hand, that the observed inter-limb frequency dissociation in these pathological conditions could be explained by the simultaneous implication of different central oscillators involving parallel neural pathways. On the other hand, in psychogenic tremor, the motor system should be intact and rhythmic contraction could be mediated by top-down synchronization involving a common oscillator system at a higher cerebral level. The ability to drive tremor frequency is recognized as characteristic of psychogenic tremor. When asked to tap out a beat with the limb contralateral to the tremulous limb under examination, tremor in the latter dissipates or more specifically shifts to the frequency of tapping (Fig. 1) or is replaced by irregular jumpy movements [16,19]. When very marked, frequency dissociation and frequency driving can be determined clinically, but most of the cases are clearly assessed only by tremor recording. Tremor recording also provides objective documentation of this phenomenon. Frequency driving can be elicited in tremors of all the body parts including lower limb or trunk. Preferably, it should be paced using a metronome and should test different consecutive frequencies, to avoid using a leading frequency that is harmonic to the abnormal movement studied. By contrast, in organic tremor, the voluntarily paced movement has no influence on the frequency of the abnormal movement. In the latter case, one can observe two simultaneous different frequencies in two different limbs, devoid of harmonic link or temporal coupling, as assessed by coherence studies. This test is thought to be the more discriminating in differentiating dystonic tremor from psychogenic tremor [17].
Paradoxical increase of tremor frequency and amplitude with mass loading An increase of tremor amplitude with mass loading is strongly suggestive of psychogenic tremor, since 70% of the patients display this feature [5], but this sign is not specific. Indeed, it may also be seen in essential tremor and parkinsonian tremor (20% and 5% of the patients, respectively), particularly in severe tremors. More interestingly, a paradoxical increase of both amplitude and frequency or a change of the location (i.e. from proximal to distal part of the limb or from one limb to another) or a sudden shift of the direction (i.e. from flexion/extension to pronation/supination) of the tremor should be searched for, as good clues for psychogenic tremor.
Co-contraction preceding tremor onset Co-activation of the antagonistic muscles of the tremulous joint is frequent in psychogenic tremor [5]. It can be detected on clinical examination and must be searched for on polymyographic recordings, if tremor onset is captured. The co-activation is thought to allow the triggering of tremor initiation. By contrast, organic tremor occurs without such co-activation, as for the typical example of parkinsonian rest tremor.
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
4
E. Apartis
Figure 1 Polymyographic recording showing a positive entrainment test (A) in a patient affected by a unilateral psychogenic tremor (right upper limb). When the contralateral upper limb is at rest (a), the spontaneous right limb tremor frequency is 5.8 Hz as shown by EMG and Acc (b). The patient is then asked to tap out a beat with the left hand at a 3.5 Hz rhythm paced by a metronome. While he is correctly performing this voluntary rhythmic movement (c), the tremor frequency switches rapidly to the voluntarily paced movement frequency, namely 3.5 Hz (d). Note that the difference of tremor frequencies measured in these two conditions (b minus d) is greater than 1.5 Hz and that these two frequencies have no harmonic link. After the test (B), the tremor keeps this new acquired frequency (3,4 Hz; (f)), while the other hand remains at rest (e). ECR: extensor carpi radialis; FCR: flexor carpi radialis; Acc: accelerometer.
Limits of the diagnostic value of distractibility and variability Variability is one of the major criteria of PMD in the Fahn and Williams classification [8]. One aspect of this variability is distractibility. Distractibility can be clearly assessed using polymyographic recordings, which may show short and sudden interruptions of the tremor or a transitory dramatic decrease of its amplitude, orchestrated by the examiner. On Acc signals, an abrupt and ‘‘squared’’ interruption of Acc oscillatory signal is obviously different from the waxing and waning modulations of the Acc amplitude typical for the parkinsonian tremor (personal observations). This is an important point, considering that rest tremor in Parkinson’s disease (PD) is mainly intermittent, and highly modulated by the emotional state and mental activation level. Moreover, one should consider carefully the diagnosis value of distractibility, particularly when the induced tremor changes do not show frequency switches or EMG pattern changes (i.e. from rhythmic to shuddering pattern).
Burst length is typically greater than 50 ms in psychogenic tremor Tremor bursts of very short duration are likely to be organic. Burst durations below 50 ms are characteristic of cortical myoclonic tremor and primary orthostatic tremor.
Influence of psychogenic tremor on contralateral motor performance In psychogenic tremor, while attempting to produce two rhythms with distant body parts, either the externally paced or the self-generated rhythm is typically poorly maintained. Bilateral recording allows study of the influence of tremor on motor performance of the contralateral limb and also conveys important information. Contralateral voluntary motor
tasks are strongly affected when tremor is present, but are preserved when the tremor has stopped. Conversely, voluntary motor tasks are not affected by the presence of an organic tremor in the opposite side. This contrast was clearly demonstrated with dual interference reaction time studies [14] and can be advantageously shown at the bedside, using bilateral polymyographic recording of auto-generated rhythm, on one hand, and tremor on the other hand, in psychogenic patients. Typically in this setting, one must focus on the quality of the contralateral voluntary rhythm paced by a metronome and compare two states: • ipsilateral limb at rest, with no involuntary tremor and; • ipsilateral limb engaged in action tremor. In this situation, a high-quality regular rhythm of the voluntary movement in state 1 compared to a disorganized movement in state 2 would provide a firm argument in favour of psychogenic tremor. A similar performance, either good or poor, in both situations has to be carefully interpreted in the light of the overall clinical picture. Pitfalls and difficulties Clonus, enhanced physiological tremor and psychogenic tremor. As described above, in psychogenic tremor, continuous voluntary-like production of tremor may be uncovered by positive frequency driving of tremor. However, some patients with PMD have tremor that cannot be altered, which would be unlikely produced via voluntary-like mechanisms. In such cases a second involuntary pathophysiological mechanism of psychogenic tremor has been evoked, considering patients with bilateral psychogenic tremor at near equal frequencies in the high range (7—10 Hz), but without inter-limb coherence [23], triggered by co-contraction. In these cases, a careful clinical examination to detect pyramidal signs is particularly important to avoid misdiagnosis of clonus. Two well-known oscillations in this frequency band and occurring independently in different limbs are clonus [24], at
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders 8 Hz and 6 Hz, in the upper and lower limbs [12] and physiologic tremor at 8—12 Hz [6]. Physiological tremor may be enhanced by anxiety, but we would not consider this per se to be a psychogenic tremor. It is characterized by a stable frequency in an individual subject, classically without demonstrable EMG rhythmic activity in forearm muscles, and devoid of clinical co-activation signs or distractibility. Moreover, in contrast to psychogenic tremor, enhanced physiological tremor has no influence on motor performance. Psychogenic parkinsonism, dystonic tremor and confounding associations. In cases of possible psychogenic parkinsonism, the comparison of clinical signs, neurophysiological features and dopamine transporter imaging results is necessary [1] to improve diagnostic accuracy. It is also important to note that 25—30% of patients with PMD have a coexisting non-psychogenic movement disorder or other neurological disease [7]. The co-occurrence of both psychogenic tremor and organic tremor is not uncommon. Patients can readily enhance and embellish the symptoms of the underlying organic disease to form a florid clinical picture and the neurophysiological recording may help sort out both components. Finally, the distinction between psychogenic and dystonic tremor may remain a diagnostic challenge. Primary orthostatic tremor and psychogenic orthostatic tremor. Primary orthostatic tremor (POT) is characterized by a fast rhythmic activity (frequency range: 14—18 Hz) that occurs when patients are standing still and disappears at rest. Patients are often misdiagnosed as having a psychogenic disorder because their main complaint is unsteadiness during orthostatism that paradoxically disappears when walking. Tremor is an infrequent symptom and seldom detected at clinical examination. Polymyographic recordings definitively demonstrate the specific high-frequency EMG pattern [11]. Orthostatic psychogenic tremor can be easily distinguished from POT on this frequency criterion (see above). Other causes of orthostatic tremor can also be demonstrated by neurophysiological examination, such as orthostatic myoclonus [9], slow orthostatic tremor [15], and cerebellar deficit or dystonic tremor. In addition other signs described above must be looked for, namely frequency driving, frequency variability and EMG pattern instability. Sometimes in complex or borderline cases, recording sessions must be repeated to assess the variability of the tremor.
A therapeutic role for the neurophysiological recording session In some cases, visualisation of the neurophysiological data can help the physician guide the patient towards the understanding of his symptoms and open a window towards recovery. In psychogenic tremor, many patients can appreciate the distractibility of their movement disorder when it is pointed out to them. An objective demonstration, based on physiological data and patient insight during the tests, allows the examining doctor to deliver an explicit statement of belief that the physical signs and symptoms are real, but positively different from signs seen in organic disease. It could be very useful in persuading the patients of the accuracy of the diagnosis and of the potential reversibility of their symptoms [26].
5
Myoclonus Generally speaking, neurophysiological study of myoclonus requires techniques exploring cortical excitability, namely conventional EEG, EEG Jerk-Locked-Back-Averaging (JLBA), somato-sensory evoked potentials (SEP), and C-reflex studies in addition to EMG. Importantly, polymyographic analysis is critical to identify neurophysiologic clues of psychogenic jerks but should usually be completed by other potentially decisive approaches. Indeed, and more specifically to PMD, pre-movement potential recording and measures of the stimulus-induced jerks latencies may help differentiate psychogenic jerks from organic myoclonus. However, the clinical presentation of psychogenic myoclonus is extremely rich and polymorphous and can mimic virtually all forms of cortical, subcortical of spinal myoclonus. Focal, multifocal, erratic, axial or generalized jerks can occur. Psychogenic jerks can be sporadic or repetitive, rhythmic or arrhythmic, spontaneous or induced by different kinds of stimuli. Analysis of all these features is critical in formulating an individualized neurophysiological strategy (Fig. 2).
Role of polymyography in evaluating psychogenic jerks To study myoclonus, the core analysis of polymyography comprises the recognition of a repeatable recruitment pattern of the jerks. Cortical, reticular and propriospinal myoclonus have a stereotyped pattern of propagation along the different muscle territories, according to the location of the hyperexcitable area in the central nervous system and depending upon the neural pathways that convey the abnormal excitation. The lack of a recognizable and reproducible recruitment pattern (Fig. 3) of spontaneous or stimuli-induced jerks provides a powerful argument for a psychogenic origin. One major difficulty, however, remains in the distinction of psychogenic jerks from dystonic myoclonus, which likewise is not organized and can misleadingly be considered as partially or paradoxically distractible. Importantly, clinical and neurophysiological features must be taken together to ensure accurate diagnosis. Nevertheless, two kinds of reproducible patterns may be encountered in psychogenic jerks. Firstly, the EMG pattern of psychogenic jerks in a limb is often organized in a triphasic sequence of agonistic/antagonistic muscle activation, as occurs during rapid voluntary ballistic movements [28]. The duration of bursts of voluntary muscle activity in a ballistic movement is generally within the range of 75—150 ms, depending on the amplitude and speed of movement. By contrast, cortical and reticular reflex myoclonus produces brief bursts (20—50 ms) of muscle activities, recorded simultaneously in antagonist muscles [25]. Secondly, a distinctive pattern of EMG activity in the axial muscles suggesting the slow spread of excitability from midthoracic region up and down the spinal cord, defined the concept of a propriospinal myoclonus [3]. As this EMG pattern can be mimicked voluntarily [13], patients displaying axial jerks, with these neurophysiologic features may be psychogenic
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
6
E. Apartis
Figure 2
Individualized neurophysiological strategy, according to the clinical presentation of the assumed psychogenic jerks.
Figure 3 Polymyography of a patient displaying generalized psychogenic jerks. Clinically, the core of his jerks consists in spontaneous bilateral extension of the upper limbs and trunk, occurring at rest, in the sitting or lying position. The diagnosis of tics is ruled out as these jerks cannot be controlled voluntarily and are not preceded by an ‘‘urge to move’’ perception. On polymyography, there is no recognizable propagation pattern that would be suggestive of a spinal, reticular or cortical source of these long duration bursts (200—800 ms). No short-delay cortical transient preceding the jerks over the corresponding central regions is recorded (curves not shown). A strong argument for the psychogenic nature of the movement is notably the lack of clear reproducibility of the order (1—5) of muscular recruitment in consecutive jerks recorded on the right side. The corresponding BP analysis is shown in Fig. 4. SCM: sternocleidomastoid muscle; FDI: first dorsal interosseous muscle; T9-T10 RA: rectus abdominis muscle at the T9-T10 level; VL: vastus lateralis muscle; TA: tibialis anterior muscle.
[32] and should be carefully evaluated for other testing criteria. As developed above, the ability to drive tremor frequency is recognized as characteristic of psychogenic tremor. Similarly, competitive rhythmic tasks may also be appropriate to modify the frequency of rhythmic psychogenic myoclonus that cannot be assessed with BP recordings, due to their frequency being too high.
Pre-movement potential or bereitschaftspotential (BP) Cortical events preceding voluntary movement can be recorded over cortical central and frontal electrodes, using the EEG jerk-locked-back-averaging (JLBA) method. In normal subjects, a self-initiated movement is preceded by a slow negative potential shift starting at about 1s before
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders
7
Figure 4 Bereitschaftspotential (BP) recordings obtained from one patient with generalized psychogenic jerks (a,b) (corresponding polymyography shown in Fig. 3) and one patient with propriospinal myoclonus (PSM) (c, d) during spontaneous myoclonus or during a self-paced voluntary movement. The averaging is triggered by the spontaneous jerks recorded on the right rectus abdominis at T9-T10 level (a,c) or by the voluntary movement obtained from the right extensor carpi radialis radialis muscle (b,d). Note over the midline central area Cz, the lack of BP preceding PSM (c) and the negative slopes preceding voluntary movement onset by 860 ms (b) and 910 ms (d), and the psychogenic jerk by 840 ms (a).
the movement onset. BP reflects preparatory activities arising from primary sensori-motor cortex and supplementary motor area (SMA). This method may be helpful in order to distinguish voluntary movements from involuntary movements. Mimicked voluntary jerks are preceded by a pre-movement potential, [18] whereas organic myoclonus are not [29]. On the contrary, spontaneous psychogenic jerks are classically accompanied by a BP (Fig. 4) [27,29]. Remarkably, BP seems to be frequently lacking prior to intended wrist extension in PMD patients [29] in parallel to impaired execution of voluntary tasks in psychogenic tremor [30], which possibly reflects an attentional bias towards the psychogenic movement disorder. Simple motor tics may be preceded by a truncated BP [2,18], which starts later compared with a BP preceding a self-paced movement. Some methodological limitations must be considered. The poor signal-to-noise ratio requires the averaging of at least 40 trials. Accordingly, when movements are infrequent, the value of this technique is limited particularly if there are time limitations. On the contrary, averages are unreliable when movements occur more than every 3 seconds and if many facial and head movements occur [4]. However, the presence of normal BP preceding spontaneous jerks is a strong argument for a psychogenic motor symptom.
However, an important warning must be considered as in normal subjects, the bereitschaftspotential is lacking or truncated before voluntary movements triggered by an external stimulus [20], with a greater role of the premotor cortex in motor preparation, compared to SMA in this setting. Consequently, the BP cannot be used to assess reflex psychogenic jerks and care must be taken to avoid cues that might form the trigger for apparently spontaneous jerks in psychogenic myoclonus and tics [4].
Latency of stimulus-induced jerks Measuring the latency of a stimulus-induced reflex jerk is very helpful to differentiate reflex myoclonus occurring in response to various stimuli (auditory, tendon tap, touch and pinprick) from a voluntary reaction. Reflex delays greater than 100 ms and variability in response latencies are suggestive of a voluntary reaction to the stimulus and, by extension, of a psychogenic jerk [4]. For example, voluntary reaction times of imitated generalized jerks ranges were 110 to 163 ms after an auditory stimulus, a tap knee or a shock to a median nerve at wrist [28]. By contrast, transcortical response latencies characteristic of cortical myoclonus are within the range of 45—50 ms after median nerve stimulation [25] and brainstem response latencies are within the
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
8
E. Apartis
range of 40—90 ms, after supra-orbitary nerve stimulation in hyperekplexia, which is a pathological and nonhabituating enhancement of the startle response to sound or tap to the mantle region [10]. In addition, the presence of giant somato-sensory-evoked potentials and of a short latency cortical transient preceding EMG burst by 20—35 ms, points towards cortical myoclonus [25] since these are not found in psychogenic jerks.
Conclusion PMD is an important clinical problem, usually necessitating extensive and repeated investigations. Clinical neurophysiological testing can be helpful in diagnosing psychogenic myoclonus and tremor, possibly leading to definite diagnosis by the demonstration of objective clues, and finally offering an opportunity to modify the level of the awareness of the symptom. In other types of PMD, such as dystonia or spasms, further development of specific neurophysiological approaches are needed.
Disclosure of interest The author declares that he has no conflicts of interest concerning this article.
References [1] Benaderettte S, Zanotti-Fragonara P, Apartis E, Nguyen C, Trocello JM, Remy P, et al. Psychogenic parkinsonism: a combination of clinical, electrophysiological and [123I] -FP-CIT SPECT scan explorations improves diagnostic accuracy. Mov Disord 2006;21(3):310—7. [2] Berardelli A, Currà A, Fabbrini G, Gilio F, Manfredi M. Pathophysiology of tics and Tourette’s syndrome. J Neurol 2003;250(7):781—7 [Review]. [3] Brown P, Thompson PD, Rothwell JC, Day BL, Marsden CD. Axial myoclonus of propriospinal origin. Brain 1991;114:197—214. [4] Brown P, Thompson P. Electrophysiological aids to the diagnosis of psychogenic jerks, spasms and tremor. Mov Disord 2001;16(4):595—9. [5] Deuschl G, Köster B, Lücking CH, Scheidt C. Diagnostic and pathophysiological aspects of psychogenic tremors. Mov Disord 1998;13:294—302. [6] Elble RJ, Randall JE. Motor-unit activity responsible for 8- to 12-Hz component of human physiological finger tremor. J Neurophysiol 1976;39:370—83. [7] Factor SA, Podskalny GD, Molho ES. Psychogenic movement disorders: frequency, clinical profile and characteristics. J Neurol Neurosurg Psychiatry 1995;59:406—12. [8] Fahn S, Williams DT. Psychogenic dystonia. Adv Neurol 1988;50:431—55. [9] Glass GA, Ahlskog JE, Matsumoto JY. Orthostatic myoclonus: a contributor to gait decline in selected elderly. Neurology 2007;68(21):1826—30. [10] Hayes M, Thompson PD. Psychogenic movement disorders. In: Hallet M, editor. Handbook of clinical neurophysiology, Vol. 1. 2003. p. 629—39 [chapter 39]. [11] Heilman KM. Orthostatic tremor. Arch Neurol 1984;41(8): 880—1.
[12] Iansek R. The effects of reflex path length on clonus frequency in spastic muscles. J Neurol Neurosurg Psychiatry 1984;47:1122—4. [13] Kang SY, Sohn YH. Electromyography patterns of propriospinal myoclonus can be mimicked voluntarily. Mov Disord 2006;21(8):1241—4. [14] Kumru H, Begeman M, Tolosa E, Valls-Sole J. Dual task interference in psychogenic tremor. Mov Disord 2007;22(14):2077—82. [15] Leu-Semenescu S, Roze E, Vidailhet M, Legrand AP, Trocello JM, Cochen V, et al. Myoclonus or tremor in orthostatism: an under-recognized cause of unsteadiness in Parkinson’s disease. Mov Disord 2007;22(14):2063—9. [16] McAuley JH, Rothwell JC, Marsden CD, Findley LJ. Electrophysiological aids in distinguishing organic from psychogenic tremor. Neurology 1998;50:1882—4. [17] McAuley J, Rothwell J. Identification of psychogenic, dystonic, and other organic tremors by a coherence entrainment test. Mov Disord 2004;19(3):253—67. [18] Obeso JA, Rothwell JC, Marsden CD. Simple tics in Gilles de la Tourette syndrome are not refaced by a normal premovement EEG potential. J Neurol Neurosurg Psychiatry 1981;44:735—8. [19] O’Suilleabhain PE, Matsumoto JY. Time frequency analysis of tremors. Brain 1998;121:2127—34. [20] Papa SM, Artieda J, Obeso JA. Cortical activity preceding selfinitiated and externally triggered voluntary movements. Mov Disord 1991;6:217—24. [21] Peters M. Simultaneous performance of two motor activities: the factor of timing. Neuropsychologia 1977;15:461—5. [22] Raethjen J, Lindeman M, Schmaljohann H, Wenzelburger R, Pfister D, Deuschl G. Multiple oscillators are causing parkinsonian and essential tremor. Mov Disord 2000;15:84—94. [23] Raethjen J, Kopper F, Govindan RB, Volkman J, Deuschl G. Two different pathogenetic mechanisms in psychogenic tremor. Neurology 2004;63:812—5. [24] Rossi A, Mazzocchio R, Scarpini C. Clonus in man: a rhythmic oscillation maintained by a reflex mechanism. Electroencephalogr Clin Neurophysiol 1990;75:56—63. [25] Shibasaki H. Electrophysiological studies of myoclonus. Muscle Nerve 2000;23:321—35. [26] Stone J, Edwards M. Trick or treat? Showing patients with functional (psychogenic) motor symptoms their physical signs. Neurology 2012;79:282—4. [27] Terada K, Ikeda A, Van Ness PC, Nagamine T, Kaji R, Kimura J, et al. Presence of Bereitschaftspotential preceding psychogenic myoclonus: clinical application of jerk-locked back averaging. J Neurol Neurosurg Psychiatry 1995;58(6):745—7. [28] Thompson, et al. Voluntary stimulus-sensitive jerks and jumps mimicking myoclonus or pathological startle syndromes. Mov Disord 1992;7(3):257—62. [29] Van der Salm S, Tijssen M, Loelman J, van Rootselaar AF. The Bereischaftspotential in jerky movement disorders. J Neurol Neurosurg Psychiatry 2012;83:1162—7. [30] Van Poppelen D, Saifee TA, Schwingenschuh P, et al. Attention to self in psychogenic tremor. Mov Disord 2011;26: 1575—6. [31] Williams DT, Ford B, Fahn S. Phenomenology and psychopathology related to psychogenic movement disorders. In: Weiner WJ, Lang AE, editors. Behavioural neurology in movement disorders. New York: Raven Press; 1995. p. 231—57. [32] Williams DR, Cowey M, Tuck K, Day B. Psychogenic propriospinal myoclonus. Mov Disord 2008;23(9):1312—3. [33] Zeuner KE, Shoge RO, Goldstein SR, Dambrosia JM, Hallett M. Accelerometry to distinguish psychogenic from essential or parkinsonian tremor. Neurology 2003;61:548—50.
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
ARTICLE IN PRESS
NEUCLI-2402; No. of Pages 8
Neurophysiologie Clinique/Clinical Neurophysiology (2014) xxx, xxx—xxx
Disponible en ligne sur
www.sciencedirect.com
REVIEW / MISE AU POINT
Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus Neurophysiologie clinique des mouvements anormaux psychogènes : tremblement et myoclonies E. Apartis Inserm-UPMC UMRS 975-CRICM, Department of Physiology, Saint-Antoine Hospital, AP—HP, 184, rue du Faubourg Saint-Antoine, 75012 Paris, France Received 17 July 2013; accepted 25 August 2013
KEYWORDS Movement disorders; Psychogenic; Tremor; Myoclonus; Reflex; Movement preparation; Polymyography; Frequency
Summary Tremor and myoclonus are very common manifestations of psychogenic movement disorders (PMD). In this context, recording of movement disorders aims to provide objective criteria for a positive diagnosis of PMD, independently of the psychological situation. Neurophysiological observations are therefore considered to have a huge impact both on diagnosis and on therapeutic approaches. A specific recording strategy should be employed whenever the medical history or clinical clues raise the eventuality of a PMD. Polymyography coupled to accelerometry is used to demonstrate the major electrophysiological criteria of psychogenic tremor, namely spontaneous variability of tremor frequency and frequency entrainment induced by contralateral rhythmic tasks. Other features, such as paradoxical increase of tremor amplitude with mass loading, co-activation preceding tremor onset and alteration of voluntary contralateral motor performances when tremor is present, are also of interest. The clinical presentation of psychogenic myoclonus is extremely rich and polymorphous and can mimic virtually all forms of cortical, subcortical or spinal myoclonus. Focal, multifocal, axial or generalized jerks can occur. Psychogenic jerks can be sporadic or repetitive, rhythmic or arrhythmic, spontaneous or stimulus-induced. All of these parameters are crucial to determine an individualized neurophysiological strategy. Polymyography is critical to identify a ballistic pattern or a discordant or non-reproducible temporo-spatial organisation of the jerks, but has usually to be completed by other potentially decisive approaches. Reflex psychogenic myoclonus for example displays stimulus-response delays that are too long and variable. Spontaneous psychogenic jerks may be also preceded by a pre-movement potential, detectable by jerk-locked-back-averaging methods. © 2013 Elsevier Masson SAS. All rights reserved.
E-mail address: [email protected] 0987-7053/$ – see front matter © 2013 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.neucli.2013.08.014
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
2
E. Apartis
MOTS CLÉS Mouvements anormaux ; Psychogène ; Tremblement ; Myoclonies ; Préparation du mouvement ; Polymyographie ; Fréquence ; Réflexe
Résumé Les mouvements anormaux psychogènes prennent très fréquemment la forme de tremblements ou de myoclonies. L’objectif de l’enregistrement des mouvements anormaux est d’apporter des arguments objectifs de diagnostic positif, indépendamment du contexte psychologique. Les observations neurophysiologiques peuvent avoir un impact important sur l’issue diagnostique et la démarche thérapeutique. Une stratégie spécifique d’examen doit être mise en œuvre dès que la nature psychogène du mouvement est évoquée sur des indices cliniques. La polygraphie musculaire couplée à l’accélérométrie est l’examen de base qui permet de démasquer les deux critères électrophysiologiques majeurs de tremblement psychogène, que sont la variabilité spontanée de fréquence et l’entraînement de fréquence induit par des tâches motrices compétitives rythmiques, ainsi que l’augmentation d’amplitude sous l’effet du poids, le signe de co-activation et l’altération des performances motrices controlatérales en présence du tremblement. La présentation clinique des myoclonies psychogènes est riche et polymorphe et peut mimer différentes formes de myoclonies, corticales, sous-corticales ou spinales. Les myoclonies psychogènes peuvent être focales, multifocales, axiales ou généralisées ; leur survenue peut être sporadique ou répétitive, rythmique ou arythmique, spontanée ou induite par les stimulations. Tous ces paramètres sont essentiels pour déterminer une stratégie neurophysiologique individualisée. La polymyographie est essentielle pour identifier un pattern balistique ou une organisation temporo-spatiale discordante ou non reproductible, mais doit être généralement complétée par d’autres approches. Les myoclonies psychogènes réflexes sont caractérisées par des délais de réponse aux stimulations longs et variables. La présence d’un potentiel de préparation au mouvement apporte un argument en faveur de la nature psychogène de myoclonies spontanées. © 2013 Elsevier Masson SAS. Tous droits réservés.
Introduction Psychogenic movement disorders (PMD) are common. As highlighted by Brown and Thompson [4], fifty percent of patients with a PMD improve to some degree and only onethird resolve. These patients with better outcome tend to have a shorter duration of movement disorder, suggesting that early diagnosis and treatment may be important. Clinical features suggestive — but not diagnostic — of PMD are the sudden onset of an inconsistent and variable movement disorder, which is incongruous with typical ‘‘organic’’ movement disorder. Other clues are marked reduction of the abnormal movement with distraction, dramatic increase in severity and complexity of movement induced by direct observation and the occurrence of spontaneous periods of remission. The evidence of underlying psychopathology and associated psychogenic symptomatology is also suggestive, but not necessarily diagnostic of a psychogenic etiology. More convincing, if detected, is the disappearance of the movement disorder when supposedly unobserved or following suggestion or placebo [4,31]. The neurophysiological examination is regularly crucial in diagnosing PMD, assisting the clinician in evaluating these patients. In this setting, the recording of movement disorders aims to provide objective criteria for a positive diagnosis of PMD, by using measurable features, independently of the psychological context. Neurophysiological observations thus have a huge impact on both diagnosis and therapeutic approaches. Additionally, in some cases, visualisation of neurophysiological data helps the physician guide the patient towards better understanding of his symptoms, opening a window towards recovery. A specific recording strategy should be performed whenever the medical history or clinical clues raise the possibility of a PMD or when the electrophysiological criteria of an
organic abnormal movement cannot be identified. As clinical neurophysiological testing can be helpful mainly in diagnosing psychogenic myoclonus and tremor, and can lead to definite diagnosis, we will focus on both these aspects in this review.
Psychogenic tremor Tremor is a frequent manifestation of PMD. After a reminder of the clinical clues of psychogenic tremor, the neurophysiological tools that should be used to reveal objective positive signs in comparison to the classical characteristics of tremors of organic origin will be detailed in the following section. Psychogenic tremor should be distinguished chiefly from pyramidal clonus and organic tremors such as enhanced physiologic tremor, parkinsonian tremor, dystonic tremor, and odd tremors, in order to allow optimal management. Aside general clinical features, relevant criteria revealed by electrophysiological study in psychogenic tremor are: • spontaneous variability of tremor frequency; • frequency driving induced by contralateral rhythmic tasks; • paradoxical increase of tremor frequency and amplitude with mass loading, and; • co-contraction preceding tremor onset (co-activation sign). By contrast, organic tremors have a stable frequency that cannot be driven to another one and are classically amplified by contralateral tasks. When clinical red flags are present, a specific recording protocol should be proposed in order to unmask these features. This is based on simultaneous
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders recording of the limb concerned by the abnormal movement with an accelerometer (Acc) and surface electromyography (EMG) of a few relevant muscles, which allows both measurement of tremor frequency and EMG burst duration, and study of the pattern of contraction between antagonist muscles. Simultaneous recording of the contralateral limb is critical to demonstrate the frequency driving induced by a variety of manoeuvres such as finger tapping or other repetitive movements in response to an external signal from a metronome. Methods of signal analysis, including timefrequency analysis of EMG and inter-limb EMG coherency are helpful to answer these questions. The former allows better definition of the frequency content of the tremor while the latter demonstrates the temporal link between two distant rhythmic activities. When a high coherency level evidences such a temporal link in two different limbs, one can hypothesize that both rhythmic activities have a common generator in the brain or depend on two coupled systems. For example, when a control subject voluntarily drums to the beat at the same frequency with both hands, right-left coherency is high. By contrast, there is no right-left coherency in essential or parkinsonian tremors, since in these situations two different central uncoupled oscillators are involved for different tremulous limbs [22]. Let us now describe more precisely what are classically considered as neurophysiological signs leading to a positive diagnosis of psychogenic tremor.
Tremor frequency and its spontaneous variability The frequency of psychogenic tremor is comprised between 4 and 10 Hz [5], and there is a large overlap between psychogenic tremor and organic tremor frequencies, except for high-frequency (14—18 Hz) orthostatic tremor. Low frequency (3 Hz) psychogenic tremors are rare, and this frequency is more evocative of mesencephalic or cerebellar tremors. The upper frequency limit of voluntary mimicked tremor is 11 Hz, so that any tremor that is faster than 12 Hz is very likely to have an organic aetiology [4]. As it is very difficult to maintain a voluntary oscillatory limb movement at the same frequency for any length of time, frequency variability in the same limb segment and in the same postural condition is strongly suggestive of a psychogenic tremor. Zeuner et al. [33] proposed 1.5 Hz as the lower limit of this frequency variability in psychogenic tremor. When variability is not spontaneously present, it can be elicited by mental or contralateral motor manoeuvres, the aim of which is to distract the patient. These manoeuvres should preferably be complex, in order to sufficiently engage attentional resources.
Influence of contralateral rhythmic tasks on tremor and inter-limb frequency dissociation An important sign of psychogenic tremor relates to the difficulty in voluntarily maintaining two or more unrelated, non-harmonic, rhythms in different body parts [21]. Most people seem incapable of producing more than one bimanual frequency. Only trained musicians are able to maintain bimanual polyrhythms. Therefore, the simultaneous occurrence in separate limbs of tremors with different
3 frequencies (frequency dissociation) is highly suggestive of organic tremor [4]. Assuming that certain motor systems are disrupted in Parkinson’s disease and essential tremor, it is proposed, on one hand, that the observed inter-limb frequency dissociation in these pathological conditions could be explained by the simultaneous implication of different central oscillators involving parallel neural pathways. On the other hand, in psychogenic tremor, the motor system should be intact and rhythmic contraction could be mediated by top-down synchronization involving a common oscillator system at a higher cerebral level. The ability to drive tremor frequency is recognized as characteristic of psychogenic tremor. When asked to tap out a beat with the limb contralateral to the tremulous limb under examination, tremor in the latter dissipates or more specifically shifts to the frequency of tapping (Fig. 1) or is replaced by irregular jumpy movements [16,19]. When very marked, frequency dissociation and frequency driving can be determined clinically, but most of the cases are clearly assessed only by tremor recording. Tremor recording also provides objective documentation of this phenomenon. Frequency driving can be elicited in tremors of all the body parts including lower limb or trunk. Preferably, it should be paced using a metronome and should test different consecutive frequencies, to avoid using a leading frequency that is harmonic to the abnormal movement studied. By contrast, in organic tremor, the voluntarily paced movement has no influence on the frequency of the abnormal movement. In the latter case, one can observe two simultaneous different frequencies in two different limbs, devoid of harmonic link or temporal coupling, as assessed by coherence studies. This test is thought to be the more discriminating in differentiating dystonic tremor from psychogenic tremor [17].
Paradoxical increase of tremor frequency and amplitude with mass loading An increase of tremor amplitude with mass loading is strongly suggestive of psychogenic tremor, since 70% of the patients display this feature [5], but this sign is not specific. Indeed, it may also be seen in essential tremor and parkinsonian tremor (20% and 5% of the patients, respectively), particularly in severe tremors. More interestingly, a paradoxical increase of both amplitude and frequency or a change of the location (i.e. from proximal to distal part of the limb or from one limb to another) or a sudden shift of the direction (i.e. from flexion/extension to pronation/supination) of the tremor should be searched for, as good clues for psychogenic tremor.
Co-contraction preceding tremor onset Co-activation of the antagonistic muscles of the tremulous joint is frequent in psychogenic tremor [5]. It can be detected on clinical examination and must be searched for on polymyographic recordings, if tremor onset is captured. The co-activation is thought to allow the triggering of tremor initiation. By contrast, organic tremor occurs without such co-activation, as for the typical example of parkinsonian rest tremor.
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
4
E. Apartis
Figure 1 Polymyographic recording showing a positive entrainment test (A) in a patient affected by a unilateral psychogenic tremor (right upper limb). When the contralateral upper limb is at rest (a), the spontaneous right limb tremor frequency is 5.8 Hz as shown by EMG and Acc (b). The patient is then asked to tap out a beat with the left hand at a 3.5 Hz rhythm paced by a metronome. While he is correctly performing this voluntary rhythmic movement (c), the tremor frequency switches rapidly to the voluntarily paced movement frequency, namely 3.5 Hz (d). Note that the difference of tremor frequencies measured in these two conditions (b minus d) is greater than 1.5 Hz and that these two frequencies have no harmonic link. After the test (B), the tremor keeps this new acquired frequency (3,4 Hz; (f)), while the other hand remains at rest (e). ECR: extensor carpi radialis; FCR: flexor carpi radialis; Acc: accelerometer.
Limits of the diagnostic value of distractibility and variability Variability is one of the major criteria of PMD in the Fahn and Williams classification [8]. One aspect of this variability is distractibility. Distractibility can be clearly assessed using polymyographic recordings, which may show short and sudden interruptions of the tremor or a transitory dramatic decrease of its amplitude, orchestrated by the examiner. On Acc signals, an abrupt and ‘‘squared’’ interruption of Acc oscillatory signal is obviously different from the waxing and waning modulations of the Acc amplitude typical for the parkinsonian tremor (personal observations). This is an important point, considering that rest tremor in Parkinson’s disease (PD) is mainly intermittent, and highly modulated by the emotional state and mental activation level. Moreover, one should consider carefully the diagnosis value of distractibility, particularly when the induced tremor changes do not show frequency switches or EMG pattern changes (i.e. from rhythmic to shuddering pattern).
Burst length is typically greater than 50 ms in psychogenic tremor Tremor bursts of very short duration are likely to be organic. Burst durations below 50 ms are characteristic of cortical myoclonic tremor and primary orthostatic tremor.
Influence of psychogenic tremor on contralateral motor performance In psychogenic tremor, while attempting to produce two rhythms with distant body parts, either the externally paced or the self-generated rhythm is typically poorly maintained. Bilateral recording allows study of the influence of tremor on motor performance of the contralateral limb and also conveys important information. Contralateral voluntary motor
tasks are strongly affected when tremor is present, but are preserved when the tremor has stopped. Conversely, voluntary motor tasks are not affected by the presence of an organic tremor in the opposite side. This contrast was clearly demonstrated with dual interference reaction time studies [14] and can be advantageously shown at the bedside, using bilateral polymyographic recording of auto-generated rhythm, on one hand, and tremor on the other hand, in psychogenic patients. Typically in this setting, one must focus on the quality of the contralateral voluntary rhythm paced by a metronome and compare two states: • ipsilateral limb at rest, with no involuntary tremor and; • ipsilateral limb engaged in action tremor. In this situation, a high-quality regular rhythm of the voluntary movement in state 1 compared to a disorganized movement in state 2 would provide a firm argument in favour of psychogenic tremor. A similar performance, either good or poor, in both situations has to be carefully interpreted in the light of the overall clinical picture. Pitfalls and difficulties Clonus, enhanced physiological tremor and psychogenic tremor. As described above, in psychogenic tremor, continuous voluntary-like production of tremor may be uncovered by positive frequency driving of tremor. However, some patients with PMD have tremor that cannot be altered, which would be unlikely produced via voluntary-like mechanisms. In such cases a second involuntary pathophysiological mechanism of psychogenic tremor has been evoked, considering patients with bilateral psychogenic tremor at near equal frequencies in the high range (7—10 Hz), but without inter-limb coherence [23], triggered by co-contraction. In these cases, a careful clinical examination to detect pyramidal signs is particularly important to avoid misdiagnosis of clonus. Two well-known oscillations in this frequency band and occurring independently in different limbs are clonus [24], at
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders 8 Hz and 6 Hz, in the upper and lower limbs [12] and physiologic tremor at 8—12 Hz [6]. Physiological tremor may be enhanced by anxiety, but we would not consider this per se to be a psychogenic tremor. It is characterized by a stable frequency in an individual subject, classically without demonstrable EMG rhythmic activity in forearm muscles, and devoid of clinical co-activation signs or distractibility. Moreover, in contrast to psychogenic tremor, enhanced physiological tremor has no influence on motor performance. Psychogenic parkinsonism, dystonic tremor and confounding associations. In cases of possible psychogenic parkinsonism, the comparison of clinical signs, neurophysiological features and dopamine transporter imaging results is necessary [1] to improve diagnostic accuracy. It is also important to note that 25—30% of patients with PMD have a coexisting non-psychogenic movement disorder or other neurological disease [7]. The co-occurrence of both psychogenic tremor and organic tremor is not uncommon. Patients can readily enhance and embellish the symptoms of the underlying organic disease to form a florid clinical picture and the neurophysiological recording may help sort out both components. Finally, the distinction between psychogenic and dystonic tremor may remain a diagnostic challenge. Primary orthostatic tremor and psychogenic orthostatic tremor. Primary orthostatic tremor (POT) is characterized by a fast rhythmic activity (frequency range: 14—18 Hz) that occurs when patients are standing still and disappears at rest. Patients are often misdiagnosed as having a psychogenic disorder because their main complaint is unsteadiness during orthostatism that paradoxically disappears when walking. Tremor is an infrequent symptom and seldom detected at clinical examination. Polymyographic recordings definitively demonstrate the specific high-frequency EMG pattern [11]. Orthostatic psychogenic tremor can be easily distinguished from POT on this frequency criterion (see above). Other causes of orthostatic tremor can also be demonstrated by neurophysiological examination, such as orthostatic myoclonus [9], slow orthostatic tremor [15], and cerebellar deficit or dystonic tremor. In addition other signs described above must be looked for, namely frequency driving, frequency variability and EMG pattern instability. Sometimes in complex or borderline cases, recording sessions must be repeated to assess the variability of the tremor.
A therapeutic role for the neurophysiological recording session In some cases, visualisation of the neurophysiological data can help the physician guide the patient towards the understanding of his symptoms and open a window towards recovery. In psychogenic tremor, many patients can appreciate the distractibility of their movement disorder when it is pointed out to them. An objective demonstration, based on physiological data and patient insight during the tests, allows the examining doctor to deliver an explicit statement of belief that the physical signs and symptoms are real, but positively different from signs seen in organic disease. It could be very useful in persuading the patients of the accuracy of the diagnosis and of the potential reversibility of their symptoms [26].
5
Myoclonus Generally speaking, neurophysiological study of myoclonus requires techniques exploring cortical excitability, namely conventional EEG, EEG Jerk-Locked-Back-Averaging (JLBA), somato-sensory evoked potentials (SEP), and C-reflex studies in addition to EMG. Importantly, polymyographic analysis is critical to identify neurophysiologic clues of psychogenic jerks but should usually be completed by other potentially decisive approaches. Indeed, and more specifically to PMD, pre-movement potential recording and measures of the stimulus-induced jerks latencies may help differentiate psychogenic jerks from organic myoclonus. However, the clinical presentation of psychogenic myoclonus is extremely rich and polymorphous and can mimic virtually all forms of cortical, subcortical of spinal myoclonus. Focal, multifocal, erratic, axial or generalized jerks can occur. Psychogenic jerks can be sporadic or repetitive, rhythmic or arrhythmic, spontaneous or induced by different kinds of stimuli. Analysis of all these features is critical in formulating an individualized neurophysiological strategy (Fig. 2).
Role of polymyography in evaluating psychogenic jerks To study myoclonus, the core analysis of polymyography comprises the recognition of a repeatable recruitment pattern of the jerks. Cortical, reticular and propriospinal myoclonus have a stereotyped pattern of propagation along the different muscle territories, according to the location of the hyperexcitable area in the central nervous system and depending upon the neural pathways that convey the abnormal excitation. The lack of a recognizable and reproducible recruitment pattern (Fig. 3) of spontaneous or stimuli-induced jerks provides a powerful argument for a psychogenic origin. One major difficulty, however, remains in the distinction of psychogenic jerks from dystonic myoclonus, which likewise is not organized and can misleadingly be considered as partially or paradoxically distractible. Importantly, clinical and neurophysiological features must be taken together to ensure accurate diagnosis. Nevertheless, two kinds of reproducible patterns may be encountered in psychogenic jerks. Firstly, the EMG pattern of psychogenic jerks in a limb is often organized in a triphasic sequence of agonistic/antagonistic muscle activation, as occurs during rapid voluntary ballistic movements [28]. The duration of bursts of voluntary muscle activity in a ballistic movement is generally within the range of 75—150 ms, depending on the amplitude and speed of movement. By contrast, cortical and reticular reflex myoclonus produces brief bursts (20—50 ms) of muscle activities, recorded simultaneously in antagonist muscles [25]. Secondly, a distinctive pattern of EMG activity in the axial muscles suggesting the slow spread of excitability from midthoracic region up and down the spinal cord, defined the concept of a propriospinal myoclonus [3]. As this EMG pattern can be mimicked voluntarily [13], patients displaying axial jerks, with these neurophysiologic features may be psychogenic
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
6
E. Apartis
Figure 2
Individualized neurophysiological strategy, according to the clinical presentation of the assumed psychogenic jerks.
Figure 3 Polymyography of a patient displaying generalized psychogenic jerks. Clinically, the core of his jerks consists in spontaneous bilateral extension of the upper limbs and trunk, occurring at rest, in the sitting or lying position. The diagnosis of tics is ruled out as these jerks cannot be controlled voluntarily and are not preceded by an ‘‘urge to move’’ perception. On polymyography, there is no recognizable propagation pattern that would be suggestive of a spinal, reticular or cortical source of these long duration bursts (200—800 ms). No short-delay cortical transient preceding the jerks over the corresponding central regions is recorded (curves not shown). A strong argument for the psychogenic nature of the movement is notably the lack of clear reproducibility of the order (1—5) of muscular recruitment in consecutive jerks recorded on the right side. The corresponding BP analysis is shown in Fig. 4. SCM: sternocleidomastoid muscle; FDI: first dorsal interosseous muscle; T9-T10 RA: rectus abdominis muscle at the T9-T10 level; VL: vastus lateralis muscle; TA: tibialis anterior muscle.
[32] and should be carefully evaluated for other testing criteria. As developed above, the ability to drive tremor frequency is recognized as characteristic of psychogenic tremor. Similarly, competitive rhythmic tasks may also be appropriate to modify the frequency of rhythmic psychogenic myoclonus that cannot be assessed with BP recordings, due to their frequency being too high.
Pre-movement potential or bereitschaftspotential (BP) Cortical events preceding voluntary movement can be recorded over cortical central and frontal electrodes, using the EEG jerk-locked-back-averaging (JLBA) method. In normal subjects, a self-initiated movement is preceded by a slow negative potential shift starting at about 1s before
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
Clinical neurophysiology of psychogenic movement disorders
7
Figure 4 Bereitschaftspotential (BP) recordings obtained from one patient with generalized psychogenic jerks (a,b) (corresponding polymyography shown in Fig. 3) and one patient with propriospinal myoclonus (PSM) (c, d) during spontaneous myoclonus or during a self-paced voluntary movement. The averaging is triggered by the spontaneous jerks recorded on the right rectus abdominis at T9-T10 level (a,c) or by the voluntary movement obtained from the right extensor carpi radialis radialis muscle (b,d). Note over the midline central area Cz, the lack of BP preceding PSM (c) and the negative slopes preceding voluntary movement onset by 860 ms (b) and 910 ms (d), and the psychogenic jerk by 840 ms (a).
the movement onset. BP reflects preparatory activities arising from primary sensori-motor cortex and supplementary motor area (SMA). This method may be helpful in order to distinguish voluntary movements from involuntary movements. Mimicked voluntary jerks are preceded by a pre-movement potential, [18] whereas organic myoclonus are not [29]. On the contrary, spontaneous psychogenic jerks are classically accompanied by a BP (Fig. 4) [27,29]. Remarkably, BP seems to be frequently lacking prior to intended wrist extension in PMD patients [29] in parallel to impaired execution of voluntary tasks in psychogenic tremor [30], which possibly reflects an attentional bias towards the psychogenic movement disorder. Simple motor tics may be preceded by a truncated BP [2,18], which starts later compared with a BP preceding a self-paced movement. Some methodological limitations must be considered. The poor signal-to-noise ratio requires the averaging of at least 40 trials. Accordingly, when movements are infrequent, the value of this technique is limited particularly if there are time limitations. On the contrary, averages are unreliable when movements occur more than every 3 seconds and if many facial and head movements occur [4]. However, the presence of normal BP preceding spontaneous jerks is a strong argument for a psychogenic motor symptom.
However, an important warning must be considered as in normal subjects, the bereitschaftspotential is lacking or truncated before voluntary movements triggered by an external stimulus [20], with a greater role of the premotor cortex in motor preparation, compared to SMA in this setting. Consequently, the BP cannot be used to assess reflex psychogenic jerks and care must be taken to avoid cues that might form the trigger for apparently spontaneous jerks in psychogenic myoclonus and tics [4].
Latency of stimulus-induced jerks Measuring the latency of a stimulus-induced reflex jerk is very helpful to differentiate reflex myoclonus occurring in response to various stimuli (auditory, tendon tap, touch and pinprick) from a voluntary reaction. Reflex delays greater than 100 ms and variability in response latencies are suggestive of a voluntary reaction to the stimulus and, by extension, of a psychogenic jerk [4]. For example, voluntary reaction times of imitated generalized jerks ranges were 110 to 163 ms after an auditory stimulus, a tap knee or a shock to a median nerve at wrist [28]. By contrast, transcortical response latencies characteristic of cortical myoclonus are within the range of 45—50 ms after median nerve stimulation [25] and brainstem response latencies are within the
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
+Model NEUCLI-2402; No. of Pages 8
ARTICLE IN PRESS
8
E. Apartis
range of 40—90 ms, after supra-orbitary nerve stimulation in hyperekplexia, which is a pathological and nonhabituating enhancement of the startle response to sound or tap to the mantle region [10]. In addition, the presence of giant somato-sensory-evoked potentials and of a short latency cortical transient preceding EMG burst by 20—35 ms, points towards cortical myoclonus [25] since these are not found in psychogenic jerks.
Conclusion PMD is an important clinical problem, usually necessitating extensive and repeated investigations. Clinical neurophysiological testing can be helpful in diagnosing psychogenic myoclonus and tremor, possibly leading to definite diagnosis by the demonstration of objective clues, and finally offering an opportunity to modify the level of the awareness of the symptom. In other types of PMD, such as dystonia or spasms, further development of specific neurophysiological approaches are needed.
Disclosure of interest The author declares that he has no conflicts of interest concerning this article.
References [1] Benaderettte S, Zanotti-Fragonara P, Apartis E, Nguyen C, Trocello JM, Remy P, et al. Psychogenic parkinsonism: a combination of clinical, electrophysiological and [123I] -FP-CIT SPECT scan explorations improves diagnostic accuracy. Mov Disord 2006;21(3):310—7. [2] Berardelli A, Currà A, Fabbrini G, Gilio F, Manfredi M. Pathophysiology of tics and Tourette’s syndrome. J Neurol 2003;250(7):781—7 [Review]. [3] Brown P, Thompson PD, Rothwell JC, Day BL, Marsden CD. Axial myoclonus of propriospinal origin. Brain 1991;114:197—214. [4] Brown P, Thompson P. Electrophysiological aids to the diagnosis of psychogenic jerks, spasms and tremor. Mov Disord 2001;16(4):595—9. [5] Deuschl G, Köster B, Lücking CH, Scheidt C. Diagnostic and pathophysiological aspects of psychogenic tremors. Mov Disord 1998;13:294—302. [6] Elble RJ, Randall JE. Motor-unit activity responsible for 8- to 12-Hz component of human physiological finger tremor. J Neurophysiol 1976;39:370—83. [7] Factor SA, Podskalny GD, Molho ES. Psychogenic movement disorders: frequency, clinical profile and characteristics. J Neurol Neurosurg Psychiatry 1995;59:406—12. [8] Fahn S, Williams DT. Psychogenic dystonia. Adv Neurol 1988;50:431—55. [9] Glass GA, Ahlskog JE, Matsumoto JY. Orthostatic myoclonus: a contributor to gait decline in selected elderly. Neurology 2007;68(21):1826—30. [10] Hayes M, Thompson PD. Psychogenic movement disorders. In: Hallet M, editor. Handbook of clinical neurophysiology, Vol. 1. 2003. p. 629—39 [chapter 39]. [11] Heilman KM. Orthostatic tremor. Arch Neurol 1984;41(8): 880—1.
[12] Iansek R. The effects of reflex path length on clonus frequency in spastic muscles. J Neurol Neurosurg Psychiatry 1984;47:1122—4. [13] Kang SY, Sohn YH. Electromyography patterns of propriospinal myoclonus can be mimicked voluntarily. Mov Disord 2006;21(8):1241—4. [14] Kumru H, Begeman M, Tolosa E, Valls-Sole J. Dual task interference in psychogenic tremor. Mov Disord 2007;22(14):2077—82. [15] Leu-Semenescu S, Roze E, Vidailhet M, Legrand AP, Trocello JM, Cochen V, et al. Myoclonus or tremor in orthostatism: an under-recognized cause of unsteadiness in Parkinson’s disease. Mov Disord 2007;22(14):2063—9. [16] McAuley JH, Rothwell JC, Marsden CD, Findley LJ. Electrophysiological aids in distinguishing organic from psychogenic tremor. Neurology 1998;50:1882—4. [17] McAuley J, Rothwell J. Identification of psychogenic, dystonic, and other organic tremors by a coherence entrainment test. Mov Disord 2004;19(3):253—67. [18] Obeso JA, Rothwell JC, Marsden CD. Simple tics in Gilles de la Tourette syndrome are not refaced by a normal premovement EEG potential. J Neurol Neurosurg Psychiatry 1981;44:735—8. [19] O’Suilleabhain PE, Matsumoto JY. Time frequency analysis of tremors. Brain 1998;121:2127—34. [20] Papa SM, Artieda J, Obeso JA. Cortical activity preceding selfinitiated and externally triggered voluntary movements. Mov Disord 1991;6:217—24. [21] Peters M. Simultaneous performance of two motor activities: the factor of timing. Neuropsychologia 1977;15:461—5. [22] Raethjen J, Lindeman M, Schmaljohann H, Wenzelburger R, Pfister D, Deuschl G. Multiple oscillators are causing parkinsonian and essential tremor. Mov Disord 2000;15:84—94. [23] Raethjen J, Kopper F, Govindan RB, Volkman J, Deuschl G. Two different pathogenetic mechanisms in psychogenic tremor. Neurology 2004;63:812—5. [24] Rossi A, Mazzocchio R, Scarpini C. Clonus in man: a rhythmic oscillation maintained by a reflex mechanism. Electroencephalogr Clin Neurophysiol 1990;75:56—63. [25] Shibasaki H. Electrophysiological studies of myoclonus. Muscle Nerve 2000;23:321—35. [26] Stone J, Edwards M. Trick or treat? Showing patients with functional (psychogenic) motor symptoms their physical signs. Neurology 2012;79:282—4. [27] Terada K, Ikeda A, Van Ness PC, Nagamine T, Kaji R, Kimura J, et al. Presence of Bereitschaftspotential preceding psychogenic myoclonus: clinical application of jerk-locked back averaging. J Neurol Neurosurg Psychiatry 1995;58(6):745—7. [28] Thompson, et al. Voluntary stimulus-sensitive jerks and jumps mimicking myoclonus or pathological startle syndromes. Mov Disord 1992;7(3):257—62. [29] Van der Salm S, Tijssen M, Loelman J, van Rootselaar AF. The Bereischaftspotential in jerky movement disorders. J Neurol Neurosurg Psychiatry 2012;83:1162—7. [30] Van Poppelen D, Saifee TA, Schwingenschuh P, et al. Attention to self in psychogenic tremor. Mov Disord 2011;26: 1575—6. [31] Williams DT, Ford B, Fahn S. Phenomenology and psychopathology related to psychogenic movement disorders. In: Weiner WJ, Lang AE, editors. Behavioural neurology in movement disorders. New York: Raven Press; 1995. p. 231—57. [32] Williams DR, Cowey M, Tuck K, Day B. Psychogenic propriospinal myoclonus. Mov Disord 2008;23(9):1312—3. [33] Zeuner KE, Shoge RO, Goldstein SR, Dambrosia JM, Hallett M. Accelerometry to distinguish psychogenic from essential or parkinsonian tremor. Neurology 2003;61:548—50.
Please cite this article in press as: Apartis E. Clinical neurophysiology of psychogenic movement disorders: How to diagnose psychogenic tremor and myoclonus. Neurophysiologie Clinique/Clinical Neurophysiology (2014), http://dx.doi.org/10.1016/j.neucli.2013.08.014
