Surface Electromyography for the Diagnosis of Tremor Syndrome: A Study of 97 Patients

Abstract

Background and Objectives:

Tremor is one of the most frequent movement disorders encountered in clinical practice with heterogeneous phenomenology and etiology. Surface electromyography (SEMG) is a noninvasive and reproducible test that can diagnose tremor syndromes.

Methods:

In this retrospective study, the clinical and electrophysiologic records of 97 consecutive patients with tremor syndromes who visited our movement disorder clinic between January 2023 and March 2024 were examined.

Results:

In our study, 28.8% (n = 28) of patients were of essential tremor (ET) syndrome. SEMG of ET syndrome patients showed synchronous bursts in 71.4% (n = 20), alternating bursts in 10.7% (n = 3), synchronous bursts with co-contraction of agonist and antagonist muscles in 10.7% (n = 3), and both synchronous and alternating bursts in 3.6% (n = 1) of patients. Tremor–ataxia syndrome formed 21.6% (n = 21) of our study population, with 71.4% (n = 15) of patients showing synchronous bursts and co-contraction and 28.6% (n = 6) patients having alternating bursts. Moreover, 12.3% (n = 12) patients were of parkinsonian tremor, of whom alternating bursts were present in 75% (n = 9) and synchronous bursts with co-contraction were present in 25% (n = 3) of patients. In addition, 11.3% (n = 11) of patients had dystonic tremor (DT), of whom 81.8% (n = 9) had synchronous bursts and co-contraction and 18.2% (n = 2) had alternating bursts.

Conclusions:

Synchronous bursts with co-contraction suggestive of DT were observed in most patients with tremor–ataxia syndrome and a small number of patients with ET syndrome. Our data suggests that SEMG helps differentiate these two clinical syndromes, which is difficult in a clinical setting.

Introduction

A tremor is an involuntary, rhythmic, and oscillatory movement involving one or several body parts.[1] It is the most common condition seen in patients who visit a movement disorder clinic.[2] The International Parkinson and Movement Disorder Society 2018 consensus guidelines on tremor classification highlight the significance of tremor phenomenology.[1] This classification gives a biaxial approach, wherein the first axis comprises history, clinical examination, and tremor characteristics giving information about the phenomenology of tremor and the second axis comprises the etiology of tremor. We often diagnose the tremor phenomenology based on clinical examination, but certain studies have shown that even the most common tremor like essential tremor (ET) is misdiagnosed in at least a third of the patients.[3] In addition to the inconsistent clinical presentation of ET, there is also disagreement among movement disorder experts over the precise definition of ET.[4] Essential tremor plus (ET plus) is defined as a tremor with the characteristics of ET with additional soft neurologic signs such as questionable dystonic posturing, impaired tandem gait, or memory impairment and it may also have an intervobserver disagreement.[1]

The postural tremor in patients with ET syndrome (ET and ET plus), dystonic tremor (DT), and tremor–ataxia syndrome (TAS) may be difficult to distinguish in certain instances.[5,6] It is very important to differentiate between them as the management of the patient changes based on the tremor phenomenology and pointers like alcohol responsiveness and the presence of family history are not always useful as a differentiating feature. In such a scenario, surface electromyography (SEMG) offers objective, reproducible, and diagnostic information about tremors, which can be complementary to clinical assessment. This helps in overcoming the inter-rater variability as well as adds objective value to the assessment. Furthermore, an objective assessment of tremors is needed especially for confirmation of tremor frequency of certain tremors like orthostatic tremor and Holme’s tremor.[1,7]

Another important use of SEMG is in functional tremors, where we can examine as well as document distractibility, entrainment, and increased amplitude of tremor upon weight loading.[8] This not only provides an objective assessment, but also improves acceptance by the patient and their family members. SEMG is also helpful in demonstrating mechanical, mechanical reflex loop, or central loop in differentiating between ET and enhanced physiologic tremor as weight loading is performed. Despite its potential utility, very few studies have been conducted on the assessment of the role of SEMG in movement disorder clinics.[9] The objective of this study was to analyze the electrophysiologic characteristics of various tremor syndromes and use SEMG as an ancillary test for studying these tremor syndromes.

Methods

This was a retrospective observational study carried out at a tertiary care teaching institute. Ethical approval was obtained from our institutional ethics committee for the retrospective data analysis. Patients who attended the movement disorder clinic at our center from January 2023 to March 2024 with the complaint of tremor and who were subsequently referred to the electrophysiology lab for SEMG test were enrolled in the study. Ninety-seven consecutively sampled patients were enrolled. The demographic and clinical data of these patients were collected from the records. The clinical diagnosis of tremor syndrome before referring the patient to the electrophysiology lab was noted as the “entry diagnosis.” Relevant history including positive family history was also noted. SEMG data was taken from electrophysiology lab records.

SEMG testing

Standardized SEMG protocol was followed. The same electrophysiology machine and set-up were used for all the patients. The software used for SEMG was “Natus Elite – Viking.” We used a six-channel system with the low-frequency filter at 0.3 Hz and high-frequency filter at 30 Hz. The agonist and antagonist muscles that were suspected of producing the rhythmic movement of the tremor were identified. A pair of electromyography (EMG) electrodes (active/recording and reference) was placed on the agonist muscle and another pair on the antagonist muscle. While assessing the upper limb tremor, extensor digitorum communis (EDC) and flexor digitorum superficialis (FDS) were the most commonly sampled muscles. When lower limb tremor was analyzed, tibialis anterior and gastrocnemius muscles were sampled, and for the assessment of neck tremors, the sternocleidomastoid and splenius capitis muscles were studied. A triaxial accelerometer was used for all the patients. Accelerometer electrodes were placed according to the axis of the motion, on a bony prominence, for example, the dorsum of the hand or the dorsum of digits to assess the tremor in the upper limbs and the dorsum of the foot while assessing the lower limb tremors. The accelerometer was placed on the vertex of the head for assessment of head tremor. Tremor frequency was assessed by Fourier analysis, based on the accelerometer findings. The tremor was recorded at rest, posture, and on action as well as terminal action, if required. For examination of postural tremor, the patient was first asked to have both arms extended forward with palms facing down and then keep the elbows flexed and palms facing toward themselves. Tremor characteristics including frequency, amplitude, burst duration, and pattern of bursts were noted. The effect of weight loading was seen after applying 500 g of weight to the wrists. Distractibility and effects of cognitive tasks on tremor characteristics were also noted. The patient was asked to tap the other hand at frequencies set by a metronome, and the presence of entrainment was checked. After reviewing these tremor characteristics, the electrophysiologic impression of the tremor was made. The clinical diagnosis and electrophysiologic characteristics were then compared.

Tremor was deemed as an enhanced physiologic tremor if the tremor frequency was between 4 and 12 Hz and the frequency decreased by >1 Hz upon weight loading due to the mechanical component of the tremor.[10]

ET syndrome was confirmed when there was a postural tremor (without any latency when switching from rest to posture), with a tremor frequency between 4 and 12 Hz, typically showing no more than a 1 Hz decrease in frequency after adding weight.[11] Tremor bursts in antagonistic muscles could be synchronous, alternating, or both.[12]

DT was diagnosed when there were synchronous bursts with co-contraction of agonist and antagonist muscles with variable bursts and an increase in tremor with certain changes in posture.[10]

The parkinsonian tremor was identified as a 4–7 Hz rest tremor (with arms at rest and hands hanging freely) showing alternating bursts on SEMG. An increase in tremor amplitude with cognitive tasks (like serial subtraction), a transient decrease in tremor after voluntary action like attainment of posture, and no change in tremor frequency after weight loading were additional pointers for parkinsonian tremor.[13,14]

Holme’s tremor was identified as a high-amplitude, low-frequency (<5 Hz), jerky tremor that was present at rest, aggravated on posture, and worsened further on action.[15]

Orthostatic tremor was labeled as a high-frequency (13–18 Hz), low-amplitude, isolated tremor syndrome that occurred when the patient was standing and subsided immediately after sitting/lying down.[1]

A diagnosis of functional tremor was made if the tremor showed features like variability (defined as a change in frequency of more than 2 Hz), suggestibility, distractibility on cognitive and ballistic tasks, and entrainment with frequencies set by a metronome.[16,17]

Statistical analysis

The presentation of categorical variables was done in the form of numbers and percentages (%). Quantitative data was presented as mean ± standard deviation (SD) and range. Data entry was done in a Microsoft Excel spreadsheet, and the analysis was done using Statistical Package for Social Sciences version 21.0.

Results

Ninety-seven patients were sampled for SEMG, which included 36% females (n = 35) and 64% males (n = 62). Of the enrolled patients, 49.4% (n = 48) were in the age group of 61–80 years [Figure 1a-d, Figure 2a-c and Table 1]. The mean ± SD of the study population was 56.2 ± 16.6 (range 13–80) years. Among the patients, 93.8% (n = 91) had tremors in the upper limbs. Lower limb and neck/head tremors were seen in 8.2% (n = 8) and 6.2% (n = 6) of the patients, respectively. EDC and FDS were the most commonly sampled muscles. The most common tremor frequency was 4–6 Hz, noted in 61.8% (n = 60) of patients. ET syndrome was the most common “entry” diagnosis after the clinical assessment, which was seen in 28.8% (n = 28) of patients. This was followed by TAS in 21.6% (n = 21) of patients, parkinsonian tremor in 12.3% (n = 12), and DT in 11.3% (n = 11) of patients.

Figure 1.

Electrophysiologic characteristics of essential tremor syndrome (a), tremor–ataxia syndrome (b), dystonic tremor (c), and parkinsonian tremor (d)

Figure 2.

Surface electromyography of tremor syndromes: (a) patient of essential tremor syndrome with synchronous bursts of 6 Hz frequency; (b) patient of parkinsonian tremor showing 5-Hz tremor with alternating bursts pattern; (c) patient of dystonic tremor showing 4-Hz tremor with synchronous burst pattern and co-contraction of agonist and antagonist muscles

Table 1.

Clinical diagnosis of patients with tremor syndrome and frequency of tremor

Referral clinical diagnosis at “entry”	No. of patients (%)	Mean tremor frequency±standard deviation (range) Hz
Essential tremor syndrome	28 (28.8%)	6.2±0.9 (4–10)
Tremor–ataxia syndrome	21 (21.6%)	5.8±1.6 (3.5–8)
Parkinsonian tremor	12 (12.3%)	5±0.6 (4–6)
Dystonic tremor	11 (11.3%)	6±1.3 (4.5–8.5)
Holme’s tremor	6 (6.2%)	3.8±1.0 (1.7–5)
Indeterminate tremor	3 (3.1%)	9.5±1.2 (8–11)
Functional tremor	3 (3.1%)	Variable (6–14)
Drug-induced tremor	3 (3.1%)	6.7±0.5 (6–7)
Enhanced physiologic tremor	3 (3.1%)	5.7±0.8 (4.6–6.6)
Cortical tremor	2 (2%)	6.75±0.8 (6–7.5)
Orthostatic tremor	1 (1%)	13
Vascular parkinsonism	1 (1%)	5
Isolated head tremor	1 (1%)	3
Restless leg syndrome	1 (1%)	5
Asterixis	1 (1%)	Variable (2–8)

ET syndrome

Twenty-eight patients were diagnosed with ET syndrome (ET + ET plus) after clinical examination. The mean ± SD age at presentation of this subset of patients was 58.5 ± 15.8 years (range 25–80) years. Of these 28 patients, 17.8% (n = 5) were clinically diagnosed with ET plus. These patients presented at a mean ± SD age of 58.8 ± 10.8 (range 47–73) years. SEMG of 71.4% (n = 20) of patients showed synchronous bursts with a tremor frequency between 5 and 10 Hz, with no change in weight loading. Of the patients, 10.7% (n = 3) showed alternating bursts and 3.6% (n = 1) had both synchronous and alternating bursts. Tremor frequency was lower (4–6 Hz) in these three patients with alternating bursts than in those with synchronous bursts. Among the patients with ET syndrome, 10.7% (n = 3, ET = 1 and ET plus = 2) had synchronous bursts in SEMG with variability in frequency and co-contraction of agonist and antagonist muscles. In addition, 3.6% (n = 1) of the ET syndrome patients had a decrease in accelerometer peak when 500 g of weight was loaded on the wrist.

Tremor–ataxia syndrome

A total of 21 patients had TAS. The mean ± SD age of these patients was 57.1 ± 12.1 (range 25–76) years. Among the patients with TAS, 71.4% (n = 15) had synchronous burst patterns in SEMG with co-contraction, favoring the dystonic component. The remaining 28.6% (n = 6) patients had alternating burst patterns.

Dystonic tremor

Eleven patients were labeled as DT after the clinical assessment. The mean ± SD age of DT patients was 56.9 ± 13.5 (range 36–75) years. Tremor frequency of these patients ranged between 4.3 and 8.5 Hz. When SEMG was performed for these 11 patients, 81.8% (n = 9) patients were found to have electrophysiologic characteristics of DT, with synchronous co-contraction of agonist and antagonist muscles. The remaining 18.2% (n = 2) patients had alternating bursts.

Parkinsonian tremor

The mean ± SD of the 12 patients clinically diagnosed with parkinsonian tremor was 56.5 ± 17.1 (range 17–79) years. SEMG characteristics of parkinsonian tremor, that is, alternating bursts, between 4 and 7 Hz frequency, were present in 75% (n = 9) patients. The remaining 25% (n = 3) of PD patients showed synchronous bursts with co-contraction on electrophysiology.

Other tremor syndromes

Of the patients, 6.2% (n = 6) had a clinical diagnosis of Holme’s tremor with a jerky, low-frequency (3–4 Hz), high-amplitude tremor on SEMG. Moreover, 3.1% (n = 3) of patients were of functional tremor and on SEMG, they showed features of distractibility with cognitive tasks and sudden motor (ballistic) tasks, variability of tremor frequency, and entrainment to frequencies set by a metronome. Other tremor syndromes included 3.1% (n = 3) of patients each with drug-induced tremor and enhanced physiologic tremor, 2% (n = 2) of patients with cortical tremor, 1% (n = 1) of patients each with orthostatic tremor, vascular parkinsonism, isolated head tremor, restless leg syndrome, and asterixis.

Discussion

Almost half of the patients in our study population were in their sixth or seventh decade at the time of presentation. Male gender was predominant in our study population (64% vs. 36%). Most (93.8%) of these patients had upper limb tremors, representing the most common region involved. Most of these upper limb tremors were distal, involving the flexor and extensor muscles. Therefore, the most commonly sampled muscles were EDC and FDS. After the clinical assessment, almost one-third of the patients were diagnosed with ET syndrome (28.8%). When electrophysiology was performed for these patients, the majority (71%) had synchronous tremor bursts and 11% had alternating tremor bursts. In previous studies conducted by Shahani and Young,[18] only 5% of patients had alternating tremor bursts on electrophysiology. Later, Sabra and Hallett[19] stated that this set of patients with alternating bursts was a subset of ET itself. In our study population, the subgroup of ET syndrome with alternating tremor bursts on electrophysiology had lower frequency (4–6 Hz) than the one with synchronous tremor bursts (5–10 Hz). These findings agreed with those observed by Deuschl et al.[12] in their study. Three patients with ET syndrome had synchronous bursts and co-contraction in SEMG. This was thought to be an electrophysiologic marker of ET plus, considering questionable dystonia as the “soft neurologic sign.”

Although ET is the most commonly made diagnosis in a patient with tremor, the presentation of ET may be heterogeneous and variable.[20] Furthermore, ET is now identified as a syndrome rather than a disease due to the possible change in the diagnosis of ET to ET plus, or as more clinical signs develop in the clinical course, the diagnosis may change to dystonia or Parkinson’s disease.[21] Patients with a clinical diagnosis of DT formed 11% of our study population. Most of these patients (82%) showed electrophysiologic characteristics of DT, that is, synchronous bursts and co-contraction.[22] There were 21 patients with TAS in our study. SEMG showed synchronous bursts and co-contraction in 15 (71%) patients with TAS. Patients presenting with postural tremors often have diagnostic confusion even after clinical assessment by a movement disorder specialist – whether they would be classified as ET plus or eventually develop a TAS. Moreover, “soft signs” of ET plus may have interexaminer variability.[23] Axis-I of the new classification of tremor provides clinical features of the tremor, but ancillary test like SEMG offers additional insights into the tremor phenomenology that clinical assessment alone might not fully capture.[1] In cases with diagnostic uncertainty, where patients may present with a bibrachial action tremor with questionable dystonia or ataxia, the etiology could be heterogenous including certain genetic causes like ANO3 and SCA12.[24] In our study, three (10.7%) patients with ET syndrome and 15 patients (71%) with TAS had synchronous bursts and co-contraction in SEMG, depicting the underlying dystonic component in both sets of patients. This gives us an understanding of the electrophysiologic overlap that exists among the three groups of patients (ET syndrome, DT, and TAS).

Other ancillary investigations include structural and receptor imaging like [18F]-dihydroxyphenylalanine positron emission tomography (DOPA-PET) scan, which can also provide valuable information about the intactness of the dopaminergic system.[25]

The third set of investigations includes genetic studies for the provisional etiology. This contributes to the axis-II of the tremor classification by providing the etiology of the tremor syndrome.[24]

Our study has some limitations. First, it is a retrospective chart review with an inherent information bias. Second, our data was collected at a tertiary care teaching institute, which may have caused a referral bias. Third, the sample size of individual tremor syndromes was small. Fourth, other biomarkers such as molecular imaging and genetics were not done in our patients.

Despite these limitations, to the best of our knowledge, our study has provided a detailed clinical and electrophysiologic profile of patients with tremor syndrome after the new tremor classification. Furthermore, our study has also provided a detailed neurophysiologic profile of patients with TAS and ET plus. The data provided by our study suggest that SEMG gives information about the tremor electrophysiologic characteristics that dictate the tremor phenomenology, and hence possibly guide patient management.

Conclusion

Synchronous bursts with co-contraction suggestive of DT were observed in most of the patients with TAS and a small number of patients with ET syndrome. Our data suggests that SEMG helps differentiate these two clinical syndromes, which is difficult in a clinical setting. SEMG should be included in the routine workup of patients with tremor syndrome as it is less expensive and provides useful information regarding the axis-I characteristics of tremor.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.