Abstract
Introduction:
Efforts to characterize the semiology of tremor have considerable value. Despite this, available data on the extent and clinical correlates of tremor asymmetry in essential tremor (ET) are surprisingly scarce. Furthermore, there are no longitudinal data.
Objective:
To enhance our understanding of tremor asymmetry in ET.
Methods:
ET cases were enrolled in a prospective, longitudinal study of older adults with ET and examined at baseline and every 18 months using a reliable and valid clinical rating scale for action tremor (range = 0 – 23 in each arm).
Results:
In most cases (263 of 306, 85.9%), there was some degree of tremor asymmetry. The range of tremor asymmetry (side-side difference in tremor scores) was 0 to 13.5. In 108 cases (35.3%, i.e., one in three cases) tremor asymmetry was ≥3 points (equivalent to at least a 1-point difference between sides on three tasks). In cross-sectional analyses, degree of tremor asymmetry was weakly inversely associated with tremor duration (Spearman’s rho = −0.154, p = 0.007). In longitudinal analyses of 89 cases with up to 4.5 years of follow-up, degree of asymmetry was consistent across time (intraclass correlation coefficients = 0.799 – 0.840, p <0.001).
Conclusions:
Most ET cases evidenced tremor asymmetry. In one third of cases, tremor asymmetry was ≥3 points (equivalent to at least a 1-point difference between sides on three tasks). In cross-sectional analyses, there was a weak inverse association between tremor asymmetry and tremor duration. During longitudinal follow-up, tremor asymmetry remained consistent, although length of follow-up was limited.
Keywords: Essential tremor, clinical, phenotype, tremor asymmetry, longitudinal
Introduction
The primary clinical feature of essential tremor (ET) is action tremor[1; 2]. Therefore, characterizing the semiology of action tremor has considerable importance, translatable in a broad range of clinical and research settings. Such knowledge should help inform clinical diagnosis, which can be particularly problematic in ET, as 30–50% of patients may be mis-diagnosed[3].
The degree of side-to-side difference in upper limb tremor severity in ET is widely regarded to be less than that in Parkinson’s disease (PD), although it is not negligible, and in some cases, the tremor may be unilateral[4; 5]. This being said, there are surprisingly few published data on side-side difference, and even fewer data of a granular nature, assessing differences in tremor severity across a range of everyday tasks. Furthermore, data on the clinical correlates of tremor asymmetry are scarce[5], and no longitudinal data on the natural history of tremor asymmetry have been published.
We leveraged data from a prospective study of ET, enrolling >300 ET cases, in which tremor severity was assessed in each arm using a reliable and valid clinical rating scale. We aimed to: (1) present granular data on tremor asymmetry across numerous daily tasks, (2) assess the clinical correlates of tremor asymmetry, and (3) provide prospective longitudinal data on tremor asymmetry across a 4.5 year time frame. The underlying a priori hypothesis was that tremor asymmetry would be apparent to a definable extent in the majority of ET cases and that the degree of tremor asymmetry might not be constant over time.
Methods
Study Design
Cases enrolled in a prospective, longitudinal, nationwide study of older adults with ET (Clinical Pathological Study of Cognitive Impairment in Essential Tremor [COGNET]; National Institutes of Health Award #R01 NS086736), described in detail[6]. The Yale University, Columbia University, and University of Texas Southwestern Medical Center Institutional Review Boards approved the study protocol. All cases provided written, informed consent.
Study Evaluations
Evaluations took place at baseline and every 18-months. At each evaluation, cases provided clinical information and a videotaped neurological examination was performed during which the severity of postural tremor and kinetic tremor (finger-nose-finger maneuver, drinking, pouring, using spoon, spiral drawing) were assessed in each arm by a movement disorders neurologist using the Washington Heights-Inwood Genetic Study of Essential Tremor (WHIGET) rating scale[7]. The scale is both reliable and valid. The scale comprises ratings of 0, 0.5, 1, 1.5, 2, 3, 4, yielding a dominant arm tremor score (range=0 – 23), a non-dominant arm tremor score (range=0 – 23), and a total Tremor Score (TTS) (range=0 – 46)[6; 8]. There is substantial to near perfect agreement between items rated with the WHIGET scale and the Tremor Research Group Essential Tremor Rating Scale (TETRAS) scale[8]. The videotaped neurological examination also included the motor portion of the Unified Parkinson’s Disease Rating Scale (UPDRS), including assessments of speech, facial expression, bradykinesia, posture, arising from a chair, and gait while walking and turning. Rest tremor was assessed with arms resting in lap and with arms at side.
Final Sample
There were 380 participants at baseline. We excluded 25 who did not have ET or who also had PD, 6 who had undergone ET brain surgery (2 unilateral deep brain stimulation, 4 bilateral deep brain stimulation, 1 unilateral focused ultrasound), and 43 with missing data. These 25 included 4 with PD. The final sample of 306 was used for cross-sectional analyses. In a subsample of 89 of these 306 cases with data from four time intervals over a 4.5 year time interval (T1 to T4) and who had not undergone ET brain surgery (this resulted in the exclusion of 9 additional cases [4 bilateral deep brain stimulation, 4 unilateral focused ultrasound, 1 bilateral focused ultrasound]), longitudinal analyses were performed.
Statistical Analyses
Normality was tested using Kolmogorov Smirnov tests; non-parametric tests were used when variables were not normally distributed. Tremor asymmetry was the difference between the dominant arm tremor score (range = 0 – 23) and the non-dominant arm tremor score (range = 0 – 23). Tremor asymmetry was also expressed as a categorical variable with three possible values: tremor score in dominant arm > tremor score non-dominant arm, tremor score in dominant arm = tremor score in non-dominant arm, or tremor score in non-dominant arm > tremor score in dominant arm. The side-side difference was the difference, when performing a particular task, between the tremor severity rating in the dominant arm and the tremor severity rating in the non-dominant arm. To assess the correlates of tremor asymmetry, we used Spearman’s correlation coefficients. We also performed stratified analyses, comparing measures of tremor asymmetry among different subgroups of ET participants. In longitudinal analyses, we used intraclass correlation coefficients to assess the agreement between tremor asymmetry at different time intervals (T1 through T4).
Results
Baseline Characteristics:
306 ET cases ranged in age from 55 to 97 (mean=79.4) years. Mean tremor duration was 36.7 years (Table 1). Similar data are presented for the 89 cases who were the subject of longitudinal analyses (Table 1); the 306 and 89 ET cases were similar with respect to demographic characteristics, family history of ET, use of ET medications, tremor severity, and tremor asymmetry.
Table 1:
Baseline tremor features of ET cases
| Demographic or clinical feature | Cross-Sectional Data | Longitudinal Data |
|---|---|---|
| n | 306 | 89 |
| Age (years) | 79.4 (8.5), range = 55 – 97 | 78.2 (9.4), range = 58 – 95 |
| Female sex | 179 (58.5) | 54 (60.7) |
| Age of tremor onset (years) | 42.5 ± 22.2 | 40.7 ± 21.4 |
| Tremor duration (years) | 36.7 ± 21.9 | 37.3 ± 22.8 |
| Family history of essential tremor | 158 (51.6) | 42 (47.2) |
| Has ever been prescribed medication for essential tremor | 253 (82.7) | 72 (80.9) |
| 5 or more | 0 (0.0) | 0 (0.0) |
| Tremor score D1 | 11.9 ± 3.6 (median = 12.0, IQR = 5.0) | 11.9 ± 3.5 (median = 12.0, IQR = 4.75) |
| Tremor score ND2 | 11.8 ± 3.7 (median = 12.0, IQR = 4.5) | 11.9 ± 3.4 (median = 12.0, IQR = 4.5) |
| Tremor asymmetry3 | 0.10 ± 3.46 (range = −13.50 – 12.00) | −0.06 ± 3.1 (range = −13.50 – 7.00) |
| Absolute value of tremor asymmetry4 | 2.48 ± 1.75 (range = 0.0 – 13.5) | 2.25 ± 2.12 (range = 0.0 – 13.5) |
| ND > D | 122 (39.9) | 41 (46.1) |
| Side-side difference >2 | 2 (0.7) | |
| Side-side difference >2 | ||
| Side-side difference >2 | 9 (2.9) | |
| Side-side difference >2 | 9 (2.9) | |
| Side-side difference >2 | 9 (2.9) | |
| Side-side difference >2 | 0 (0.0) | |
| Both arms in lap and arms at side | 12 (3.9) |
All values represent mean ± standard deviation or number (percent)
D = dominant arm. IQR = interquartile range. ND = non-dominant arm
The potential range = 0 – 23.
The potential range = 0 – 23.
Tremor score in dominant arm – tremor score in non-dominant arm. The potential range = −23 – 23
│Tremor score in dominant arm – tremor score in non-dominant arm│
This categorical variable had three possible values: tremor score D > tremor score ND, tremor score D = tremor score ND, or tremor score ND > tremor score D.
Difference, when performing this task, between the tremor severity rating in the dominant arm and the tremor severity rating in the non-dominant arm.
Cross-Sectional Analyses:
Mean and median baseline tremor scores in the dominant and nondominant arms were virtually identical (Table 1). The mean tremor asymmetry was 0.10 (i.e., on average, tremor score in the dominant arm was one-tenth of a point higher than that in the non-dominant arm, Table 1). The mean of the absolute value of mean tremor asymmetry was 2.48 (Table 1), indicating that the average side-side difference was 2.48 out of a possible 23 points. In most cases (263 or 85.9%), there was some degree of asymmetry (i.e., either the tremor score in the dominant arm was greater than that in the non-dominant arm or vice versa); in only 43 (14.1%) was the tremor score in the dominant arm equal to that in the non-dominant arm (Table 1). The range of tremor asymmetry was −13.50 to 12.00, indicating substantial asymmetry in some cases (Figure 1). In 83 (27.1%) cases, tremor asymmetry was ≤ 1 point, in 72 (23.5%) it was 1.5 to 2.5 points, and in 108 (35.3%, i.e., one in three cases) it was ≥3 points (equivalent to at least a 1-point difference between sides on three tasks). In some it was even more substantial: in 34 (11.1%) cases it was 5.5 – 10 points, and in 2 (0.7%) cases it was >10 points.
Figure 1:
Tremor asymmetry (dominant arm tremor score – non-dominant arm tremor score) in 306 ET cases. Legend: Tremor asymmetry was calculated as the difference between the dominant arm tremor score (range = 0 – 23) and the non-dominant arm tremor score (range = 0 – 23).
The degree of tremor asymmetry was weakly inversely associated with tremor duration (Spearman’s rho = −0.154, p = 0.007, i.e., the longer the tremor duration, the less apparent was the asymmetry) and weakly inversely associated with total tremor score (i.e., tremor severity) (Spearman’s rho = −0.193, p < 0.001, i.e., the greater then tremor severity, the less apparent was the asymmetry). The degree of asymmetry was not associated with age (Spearman’s rho = −0.089, p = 0.12) but was weakly associated with older age of tremor onset (Spearman’s rho = 0.132, p = 0.02). The degree of asymmetry was not associated with the number of tremor medications (Spearman’s rho = −0.03, p = 0.60).
We also performed stratified analyses, and did not find that tremor asymmetry was associated with family history of ET (Supplemental Table 1). In another stratified analysis, tremor asymmetry was not associated with younger vs. older age of tremor onset (Supplemental Table 2); this result differed somewhat from the correlation analysis above, suggesting that the difference was more readily detected when age of onset is treated as a continuous rather than a binary variable.
We also assessed asymmetry of postural tremor and kinetic tremors across a range of tasks (arm extension, finger-nose maneuver, pouring, drinking, using a spoon, drawing a spiral) (Table 1). The proportion of cases in whom asymmetry was present ranged from 42.8% (for using a spoon) to 58.8% (for pouring) (Table 1). The proportion of cases for whom there was a >1-point difference ranged from 2.65% (for finger-nose maneuver) to 17.6% (for pouring) (Table 1), with values for proportions as follows: 2.65% (finger-nose maneuver), 6.3% (postural tremor), 9.1% (spiral drawing), 9.8% (drinking), 12.7% (using a spoon), and 17.6% (pouring).
Longitudinal Analyses:
In a subsample of 89 cases with data from four time intervals over a 4.5 year time interval (T1 to T4), we assessed tremor asymmetry across time. Within a participant, there is a high level of agreement between tremor asymmetry scores across time points (Intraclass correlation coefficients ranging from 0.799 – 0.840, p < 0.001, Supplemental Table 3). Supplemental Figure 1 displays the agreement between tremor asymmetry scores at T1 and T4, the two time points that were temporally the furthest apart. Supplemental Figure 2 displays, on a participant level, that asymmetry was quite consistent over time, although there were certainly examples of individuals who switched categories (from red [positive asymmetry score] to green [negative asymmetry score] or vice versa).
Discussion
The overarching aim of this study is to enhance our understanding of tremor asymmetry in patients with ET. There were several findings.
First, in a large majority of ET cases (85.9%), there was some degree of tremor asymmetry (i.e., either the tremor score in the dominant arm was greater than that in the non-dominant arm or vice versa). The observed asymmetry of tremor aligns with our a priori hypothesis. In prior studies, the proportion with asymmetric tremor has been reported to be 39 – 46% (n = 82 patients)[9], 50% (n = 487 patients)[10], and 89% (n = 54 patients)[5], and to some extent is a function of the definition of asymmetry.
Second, the degree of tremor asymmetry ranged from −13.50 to 12.00, indicating substantial asymmetry in some cases. In one-quarter of cases, it was negligible (≤ 1 point), but in 108 (35.3%, i.e., one in three cases) it was ≥3 points (equivalent to at least a 1-point difference between sides on three tasks). In some it was even more substantial: in 34 (11.1%) cases it was 5.5 – 10 points, and in 2 (0.7%) cases it was >10 points. During specific tasks, the proportion of cases for whom there was a >1-point difference ranged from as low as 2.65% (for finger-nose maneuver) to as high as 17.6% (for pouring). In one prior study, involving 39 ET cases, the intensity of the tremor in the more affected hand was over two times higher than that in the less affected hand, which is greater than what we observed[11].
Third, in cross-sectional analyses, the longer the duration of tremor and the greater the severity of tremor, the less apparent was the asymmetry. The magnitude of these associations, however, was modest. We had hypothesized the converse, under the assumption that one side might worsen at a faster rate than the other side, therefore further widening the gap over time. However, these data suggest that a combination of factors might be operative: (1) the two sides perhaps worsen at the same rate over time although the tremor might start on one side prior to that of the other and (2) ceiling effects might reduce the magnitude of the detectable difference between the two sides over time. The combination of factors could result in an apparent diminution in tremor asymmetry with time. Our longitudinal analyses revealed that tremor asymmetry was highly consistent across time, although this is not the same as saying that tremor asymmetry was static. Furthermore, the prospective follow-up period was over a limited time frame (up to 4.5 years); longer time frames may have revealed a progressive reduction in asymmetry with time.
Fourth, overall, tremor was slightly worse in the dominant than non-dominant arm (i.e., a slightly larger number of cases had more tremor in the dominant arm). This stands in contrast to what has been observed in population-based studies, which show the converse[5]. This is to be expected. Ours was not a population-based study; cases responded to an advertisement to enroll in a research study.
There are few published studies on tremor severity. One study [5] sampled 54 community-dwelling cases with milder ET, presenting cross-sectional data on tremor asymmetry during different tasks. A recent study that utilized a smaller sample of COGNET cases (n=213) [12]presented cross-sectional data on tremor asymmetry but no granular data on tremor asymmetry during different tasks. A clinic-based study of 487 ET cases reported cross-sectional data on tremor asymmetry but no granular data on tremor asymmetry during different tasks[10]. Another cross-sectional study of 39 ET cases reported on overall tremor asymmetry[11], as did another cross-sectional clinic-based study of 82 ET cases; again more granular task-specific data were not presented[9]. No studies presented longitudinal data.
There are limitations. First, we used a clinical rating scale; accelerometry would have been more precise. However, clinical scales provide clinically-relatable measures that are more easily interpretable for a clinician. Second, we used an ordinal scale; hence intervals between tremor ranks were unequal. Third, cases who had undergone surgery were removed from both our cross-sectional and longitudinal analyses, which may have resulted in the selection of cases with a less severe phenotype. However, the number of such cases was modest. Study strengths included the large sample, use of a clinically relatable measure and a reliable and valid clinical rating scale, assessment of kinetic tremor during multiple daily activities (i.e., task-specific analyses), presentation of individual-level data, and availability of longitudinal data.
In summary, most ET cases evidenced tremor asymmetry. In one third, tremor asymmetry was ≥3 points (equivalent to at least a 1-point difference between sides on three tasks). In cross-sectional analyses there was a weak inverse association between tremor asymmetry and tremor duration. During longitudinal follow-up, tremor asymmetry remained consistent, although length of follow-up was limited.
Supplementary Material
Efforts to characterize the semiology of tremor in ET have considerable value.
Data on tremor asymmetry are surprisingly scarce and none are longitudinal.
In most cases (85.9%), arm tremor, assessed with a 0 – 4 scale, was asymmetric.
In one third, there was least a 1-point difference between sides on three tasks.
In longitudinal analyses, the degree of asymmetry was consistent across time.
Funding
This work was supported by National Institutes of Health Award #R01NS086736.
NIH played no role in the design, collection, analysis, or interpretation of the data, or in the writing of this manuscript. The authors declare that there are no conflicts of interest relevant to this work.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Ethical Compliance Statement
This study was approved by the University of Texas Southwestern Medical Center Institutional Review Board, approval numbers STU-2020–0565 and STU-2020–0564. Written informed consent was obtained from all participants.
We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.
Financial Disclosure/Conflict of Interest: Each author reports nothing to disclose. No author reports a conflict of interests.
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data Sharing:
The data that support the findings of this study are available from EDL upon reasonable request.
References
- [1].Shih LC, Essential Tremor. Continuum (Minneap Minn) 31 (2025) 979–999. [DOI] [PubMed] [Google Scholar]
- [2].Lapena-Motilva J, Sanchez-Ferro A, Mendez-Guerrero A, Rabano-Suarez P, and Benito-Leon J, Essential tremor: tips and tricks. Pract Neurol 25 (2025) 313–322. [DOI] [PubMed] [Google Scholar]
- [3].Amlang CJ, Trujillo Diaz D, and Louis ED, Essential Tremor as a “Waste Basket” Diagnosis: Diagnosing Essential Tremor Remains a Challenge. Front Neurol 11 (2020) 172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Phibbs F, Fang JY, Cooper MK, Charles DP, Davis TL, and Hedera P, Prevalence of unilateral tremor in autosomal dominant essential tremor. Mov Disord 24 (2009) 108–11. [DOI] [PubMed] [Google Scholar]
- [5].Louis ED, Wendt KJ, Pullman SL, and Ford B, Is essential tremor symmetric? Observational data from a community-based study of essential tremor. Arch Neurol 55 (1998) 1553–9. [DOI] [PubMed] [Google Scholar]
- [6].Delgado N, Berry DS, Hernandez DI, and Louis ED, Prospective, longitudinal analysis of medication use in a cohort of elderly essential tremor cases. J Neurol Sci 442 (2022) 120387. [DOI] [PubMed] [Google Scholar]
- [7].Louis ED, Ottman R, Ford B, Pullman S, Martinez M, Fahn S, and Hauser WA, The Washington Heights-Inwood Genetic Study of Essential Tremor: methodologic issues in essential-tremor research. Neuroepidemiology 16 (1997) 124–33. [DOI] [PubMed] [Google Scholar]
- [8].Hernandez AB, Berry DS, Grill N, Hall TM, Burkes A, Ghanem A, Sharma VD, and Louis ED, WHIGET and TETRAS Ratings of Action Tremor in Patients with Essential Tremor: Substantial Association and Agreement. Tremor Other Hyperkinet Mov (N Y) 14 (2024) 14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Angelini L, Paparella G, Cannavacciuolo A, Costa D, Birreci D, De Riggi M, Passaretti M, Colella D, Guerra A, Berardelli A, and Bologna M, Clinical and kinematic characterization of parkinsonian soft signs in essential tremor. J Neural Transm (Vienna) 131 (2024) 941–952. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Whaley NR, Putzke JD, Baba Y, Wszolek ZK, and Uitti RJ, Essential tremor: phenotypic expression in a clinical cohort. Parkinsonism Relat Disord 13 (2007) 333–9. [DOI] [PubMed] [Google Scholar]
- [11].Machowska-Majchrzak A, Pierzchala K, Pietraszek S, and Labuz-Roszak B, Essential tremor - assessment of tremor accelerometric parameters’ symmetry and the relationship between hand dominance and severity of tremor. Neurol Neurochir Pol 45 (2011) 121–7. [DOI] [PubMed] [Google Scholar]
- [12].Delgado N, Berry DS, and Louis ED, Distribution of rest tremor in patients with Essential Tremor: Does it lateralize with simple kinetic, postural, or intention tremors? Parkinsonism Relat Disord 102 (2022) 36–41. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from EDL upon reasonable request.