The Spiral Axis: A Comparison of Unaffected First-Degree Relatives of Essential Tremor Cases vs. Controls

Abstract

We previously observed that during a spiral drawing task, in essential tremor (ET) cases, the tremor wave forms align along a single predominant axis. This interesting clinical feature can distinguish ET from dystonia cases. We now investigate whether the unaffected relatives of ET cases also express this trait, albeit perhaps in a milder form. To address our aim, we assessed the spiral axis in 237 unaffected first-degree relatives of ET cases (FD-ET), comparing them to 105 controls (Co). A movement disorder neurologist assessed four hand drawn spirals for the presence of a single identifiable tremor orientation axis. A spiral axis score (range = 0 – 4 [a single axis on 4 spirals]) was assigned to each enrollee. FD-ET had higher spiral axis scores than Co. In a contingency table, the distribution of spiral axis scores differed in the two groups: FD-ET (highest) and Co (lowest) (ordinal chi-square test p = 0.014). Furthermore, when spiral axis scores were examined as a continuous measure, the groups differed (Mann-Whitney test p = 0.03) - with the means being 0.51 (FD-ET) and 0.26 (Co). These data have scientific implications. They (1) show that such axes are more common in relatives of ET cases than controls, and (2) raise the possibility that the spiral axis may be an early subclinical feature of ET.

1. Introduction

We previously observed that during a spiral drawing task, the wave forms of ET cases often aligned along a single predominant axis rather than several axes [1]. In an initial study of 27 ET cases with arm tremor, this clinical feature was observed in one or both arms in 23 (85.2%) ET cases [1]. In a second study, we evaluated 120 ET cases and 15 cases of dystonic tremor for the presence of this clinical feature [2]. Two right and two left hand spirals were assessed for the presence of a single identifiable tremor orientation axis, and a spiral axis score (range = 0 - 4 [a single axis on all 4 spirals]) was assigned to each case. Most (84.2%) ET cases had a spiral axis score ≥1 vs. only half (53.3%) dystonic tremor cases [2]. Since then, we have used this clinical feature, along with a host of others, to provide additional support for the diagnosis of ET [3, 4].

ET is often familial [5, 6]. Unaffected relatives of ET cases exhibit mild, subclinical tremor that is in excess of that seen in control relatives, indicating that partially expressed features of ET occur in ET families [7, 8]. Cerebellar signs and cognitive deficits, associated with ET in case-control studies [9-12], are also found to a mild degree in first-degree relatives of ET cases - mild imbalance [13], subtle cerebellar signs [14], and mild cognitive deficits [15] have been observed to a greater degree in ET relatives than control relatives.

The identification of early subclinical abnormalities among family members has value, and these abnormalities are worthy of full and detailed exploration, as they could serve as markers of individuals who are developing disease-associated traits. Some of these abnormalities could be more widespread among relatives and others less so. If and when a neuroprotective therapy were to become available for ET, these clinical tags would be of value in terms of stratifying vulnerable relatives for treatment. In the spirit of this enquiry, we sought to explore the extent to which an identifiable spiral axis may also be a subclinical feature of ET, presenting before fully manifest tremor, is unknown. The timing of presentation of this feature and whether it affects family members with only mild tremor, also remain unknown. In this study, we compared first-degree relatives of ET cases to those of controls. None of the relatives had ET. A priori, our hypothesis was that an identifiable axis would be an early marker of the ET relative group but not the other group. That is, this feature would be present to a significantly greater degree among relatives of ET cases than controls.

2. Methods

2. 1. Ascertainment of study subjects

ET cases were ascertained from study advertisements to the membership of the International Essential Tremor Foundation, members of ET research studies of the principal investigator (E.D.L.), and the clinical practice of the Yale Movement Disorders Group [13, 15]. The subject of these analyses was the unaffected first-degree relatives of ET cases (FD-ET) and age-matched controls (Co), enrolled between May 2016 and March 2020 [13, 15]. Enrollees gave written informed consent in accordance with the Declaration of Helsinki.

2.2. Screening process

ET cases, contacted by telephone, provided informed consent, and were interviewed by a trained research assistant. This interview established the case’s history of tremor (onset age, diagnosis age, tremor features). Each completed and mailed four hand-drawn spirals, provided a list of current medications, and an inventory of caffeine intake on the day spirals were drawn. During spiral drawings, enrollees drew the spiral without touching their wrist or arm to the paper [1]. Spirals were drawn freely on a standard 8.5 × 11 inch sheet of paper using a ballpoint pen while the enrollee was seated at a table [1]. The paper was centered at right angles, horizontally, directly in front of them and held down by their other hand [1]. Two spirals were drawn with each hand (4 spirals total) [16]. All materials were evaluated by E.D.L., a senior movement disorder neurologist, and diagnoses of ET assigned (i.e., moderate or greater amplitude tremor not due to another cause such as hyperthyroidism or medications). A small number of cases were assigned diagnoses of possible dystonia rather than ET, based on the following features: (1) isolated head tremor (no arm tremor), (2) irregular or jerky tremor, (3) no family history of ET, (4) endorsement of dystonia diagnosis (e.g., torticollis, writer’s cramp), and (5) no identifiable spiral axis, a feature previously shown to be more common (although not universally present) in dystonia than in ET cases.

Cases informed the investigator of all reportedly unaffected living first-degree relatives age ≥ 40 years [15]. With their permission, these relatives were contacted by telephone and screened after they were consented using a protocol approved by the Yale University Institutional Review Board. The screen included a 12-item tremor screening questionnaire [17] and questions about prior diagnoses of ET, Parkinson’s disease (PD), or dystonia [15]. They completed and mailed 4 hand-drawn Archimedes spirals (2 right, 2 left), which were rated by E.D.L. (0, 0.5, 1, 1.5, 2, and 3, see definitions and examples in Louis et al. [18]).

Relatives were initially categorized as unaffected if they met all three of the following criteria: (1) they did not report tremor during the 12-item screening questionnaire [14], (2) they had never been assigned a diagnosis of ET, and (3) their screening spirals were assigned scores <2.0 [15].

The spouses of unaffected relatives were also screened, if available. Each underwent the same screening process and were enrolled as Co if they were initially categorized as unaffected and they reported no family history of ET [15].

2.3. In-person clinical evaluation

All FD-ET and Co were invited for an in-person clinical evaluation if they were initially categorized as unaffected [15]. All were evaluated in person by a trained tester using structured clinical questionnaires to ascertain demographic, lifestyle and clinical information [15]. The Cumulative Illness Rating Scale (CIRS) (range = 0–42 [maximum co-morbidity]) [19], a measure of medical co-morbidity, was administered; this assessed both the presence and severity of comorbid medical conditions in 14 bodily systems.

Each enrollee underwent a 30 – 40-minute standardized videotaped neurological examination. This included a detailed assessment of postural tremor, five tests for kinetic tremor, the motor portion of the Unified Parkinson’s Disease Rating Scale [20] excluding an assessment of rigidity, and a comprehensive assessment of dystonia. The severity of postural and kinetic arm tremors was rated on 12 examination items using a reliable rating scale [21]. Ratings were 0, 0.5, 1.0, 1.5, 2, and 3 (for kinetic tremor); these resulted in a total tremor score (range = 0 – 36 [maximum]) [15].

FD-ET and Co were then re-evaluated for a potential ET diagnosis based on detailed review of questionnaires and videotaped neurological examination data. Diagnoses of ET were assigned based on published diagnostic criteria with established reliability [21] and validity [22] - moderate or greater amplitude kinetic tremor during three or more activities, or a head tremor in the absence of PD or another known cause [e.g., medication-induced tremor, tremor from hyperthyroidism]) [15].

2.4. Assignment of spiral axis score

Two raters, both movement disorder specialists (ME, EDL), independently assigned ratings to 20 sets of spirals with the full range of spiral axis score (0 – 4) and demonstrated very high inter-rater agreement (intraclass correlation coefficient = 0.90, p < 0.001), indicating that both raters would be highly likely to produce similar results on the full data set, thereby obviating the need for a second rater. The more senior of the two raters rated the full data set. In doing so, the rater, blinded to all clinical information, visually assessed each spiral drawing, noting the presence of tremor throughout the spiral and determining whether the tremor was most visible or most severe at certain points in the spiral and whether those points aligned along a plane or axis. Examples are provided of spirals with mild tremor and a single identifiable tremor orientation axis (Figure 1) and those mild tremor yet no single tremor orientation axis (Figure 2). Additional examples are provided in Figure 1 in Michalec et al [2]). Based on the ratings of 4 spirals per enrollee, a spiral axis score (range = 0 – 4) [no single axis on all 4 spirals - a single axis on all 4 spirals] was assigned to each enrollee [2].

Figure 1:

Four spirals with subtle tremor (marked by arrows) that lines up along a single identifiable tremor orientation axis (as marked by solid line).

Figure 2:

Four spirals with subtle tremor (marked by arrows) that is diffusely present and disorganized and does not line up along a single tremor orientation axis.

2.5. Final Sample

As detailed [14], 432 FD-ET were enrolled and after screening, we determined that there were 269 unaffected FD-ET; 237 of these had a complete set of spirals. We enrolled 191 Co, and after screening, there were 105 controls who also had a complete set of spirals.

2.6. Statistical Analyses

We first compared our groups (FD-ET vs. controls [Table 1]) in terms of demographic and clinical features. Our primary analysis was a simple one to compare the spiral axis score in FD-ET and Co, with our a priori hypothesis being that it would be higher in FD-ET.

Table 1:

Demographic and clinical characteristics of FD-ET and Controls

	FD-control (n = 105)	FD-ET (n = 237)	P value
Age (in years)	58.2 ± 10.2 [59.0]	57.2 ± 9.8 [56.0]	0.26 a
Gender [female]	35 (33.3)	168 (70.9)	< 0.001 b
Ethnicity [white]	99 (95.2)	229 (96.6)	0.55 c
Education (in years)	16.6 ± 2.9 [16.0]	16.7 ± 2.5 [16.0]	0.75 a
Current cigarette smoker	0 (0.0)	6 (2.5)	0.19 c
Cumulative Illness Rating Scale Score	3.8 ± 3.7 [3.0]	4.1 ± 3.5 [4.0]	0.22 a
Total tremor score	6.7 ± 3.0 [6.5]	7.9 ± 2.8 [7.5]	<0.001 a
Spiral axis score (Contingency table)			0.014 d
0	87 (82.9)	173 (73.0)
1	11 (10.5)	25 (10.5)
2	5 (4.8)	25 (10.5)
3	2 (1.9)	9 (3.8)
4	0 (0.0)	5 (2.1)
Spiral axis score (as a continuous measure)	0.26 ± 0.64 [0.0] IQR = 0	0.51 ± 0.97 [0.0] IQR = 1	0.03 a

All values are mean ± standard deviation [median] or number (percentage).

IQR = interquartile range.

^aMann-Whitney test.

^bChi-square test.

^cFisher’s Exact test.

^dOrdinal chi-square test.

The spiral index score was not normally distributed (Kolmogorov Smirnov test p <0.001); therefore, nonparametric tests were used when testing for association between this and other variables. Using Co (i.e., our normal group), we also assessed the correlation of spiral axis score with demographic and clinical variables.

For the primary analyses, we used two approaches to compare the spiral axis score across groups (Table 1). The first approach, which used contingency tables, assessed the proportion of each category of score across the two groups and the test statistic was an ordinal chi-square test. The second approach was treat spiral axis score as a continuous measure and to report mean and median and assess the rank order of values across groups; for this, the nonparametric test statistic was a Mann-Whitney test.

3. Results

3.1. Primary Analysis:

We compared 237 FD-ET to 105 Co (Table 1) and they were similar with respect to age, ethnicity, years of education, current smoking status, and CIRS score. A larger proportion of FD-ET were men. The total tremor score of FD-ET was higher than that of FD-controls (Table 1). In addition, 231 (98.3%) of 237 FD-ET and 101 (96.2%) of 105 Co had a spiral score of 0.5 or higher on one or more of their 4 screening spirals, reflecting the presence of some mild degree of enhanced physiological tremor in nearly all enrollees.

The spiral axis score was not correlated with age (Spearman’s r = −0.06, p = 0.55), years of education (Spearman’s r = 0.04, p = 0.73), CIRS score (Spearman’s r = −0.15, p = 0.13), or total tremor score (Spearman’s r = 0.12, p = 0.25); it did not differ by gender (Mann-Whitney p = 0.24) or white ethnicity (Mann-Whitney p = 0.94). Hence, none of these variables could have confounded our main association between diagnostic group and spiral axis score.

Using contingency tables, one may see the spiral axis score data (Table 1). Thus, in FD-controls, 87 (82.9%) did not have a predominant spiral axis on any of their 4 spirals, 11 (10.5%) had a predominant axis on 1 of their 4 spirals, 5 (4.8%) had a predominant axis on 2 of their 4 spirals, 2 (1.9%) had a predominant axis on 3 of their 4 spirals, and none had a predominant axis on all 4 of their spirals. The spiral axis scores (0, 1, 2, 3, 4) differed across the two groups, with the overall distribution of scores in the FD-ET being higher than that of Co (ordinal chi-square test p = 0.014, Table 1). Furthermore, when spiral axis score was examined as a continuous measure, FD-ET and Co differed (Mann-Whitney test p = 0.03), with the mean in FD-ET = 0.51 and the mean in Co = 0.26 (Table 1).

One issue is that the total tremor score was lowest in the Co. It is conceivable that an absence of tremor in some controls could have explained their lower overall spiral axis score because one needs some tremor in order for a spiral axis score to be visible. To explore this possibility, we excluded the 27 Co with the lowest quartile of total tremor score. In this smaller group of 78 Co, total tremor score = 7.6 ± 2.6 (median = 7.0), and the total tremor score no longer differed when compared to that of FD-ET (Mann-Whitney p = 0.29). When we compared the spiral axis score across our two groups, the difference persisted (spiral axis score in Co = 0.26 ± 0.67, median = 0; spiral axis score in FD-ET as presented in Mann-Whitney test p = 0.03). Similarly, the distribution of spiral axis scores (0, 1, 2, 3, 4) differed across the two groups, with the scores in the FD-ET being higherthan that of Co (ordinal chi-square test p = 0.03).

4. Discussion

Spiral axis score was greater in FD-ET than Co. These data have several implications. First, they show that such axes are more common in relatives of ET cases than controls. Second, they raise the possibility that the spiral axis may be an early subclinical feature of ET. The results are also in line with those of our previous studies in which we showed that features present in ET (mild imbalance [13], subtle cerebellar signs [14], and mild cognitive deficits [15]) have been observed to a greater degree in relatives of ET cases than age-matched controls.

Although FD-ET and Co differed, it is important to note that the vast majority of FD-ET (i.e., 73.0%) did not have a predominant spiral axis on any of their spirals and only 2.1% had such an axis on all four of their spirals. Hence, this feature is not only variably present but is a feature of a minority of FD-ET. It is by no means a universal finding.

One issue is that the total tremor score was slightly lower in the Co than in the other groups. It is possible that this could have accounted for their lower spiral axis score. This being said, all Co had some tremor (i.e., their total tremor score was > 0). To further explore this issue, in a sensitivity analysis, we excluded the 27 Co with the lowest quartile of total tremor score. In this smaller group of 78 Co, total tremor score = 7.6 ± 2.6 (median = 7.0, minimum score = 4), and the total tremor score no longer differed when compared to that of FD-ET (p = 0.29). Hence, now all the groups were on a level playing field with respect to tremor severity. When we compared the spiral axis score across groups, the difference persisted (p = 0.03). Similarly, the distribution of spiral axis scores differed across the groups, with the overall distribution of scores in the FD-ET being higher than that of Co (p = 0.03).

Although statistically significant, the difference between groups was subtle as our study subjects by intention and design comprised unaffected individuals who did not have a neurological disease [14]. Thus, our analysis did not lend itself to a discussion of sensitivity and specificity – we are not suggesting the presence of a certain spiral index score be used clinically to differentiate the two groups as our goal was not to develop a diagnostic differentiator [14]. Indeed, the tabular data (Table 1) show a high degree of overlap in spiral axis score between our FD-ET and Co.

One concern is whether some of our FD-ET may have had manifest ET. For several reasons, this is unlikely [15]. First, no FD-ET reported tremor during a 12-item tremor screening questionnaire [17]. Second, no FD-ET had received a diagnosis of ET by a treating physician. Third, no FD-ET met diagnostic criteria for ET or what has previously been referred to as “borderline ET” (i.e., FD-ET with borderline clinical findings who did not meet strict criteria for ET) based on a detailed neurological examination reviewed by a senior movement disorders neurologist [23]. The mean total tremor score of FD-ET was 7.9 out of 36. A score of 8 equates to a score of 1 (mild tremor) or less on each of the 12 items rated. By comparison, in prior studies of ET cases ascertained from population-based studies, the mean total tremor score was 17.8 - 19.8 [24, 25], and in genetic and clinic-based studies, the mean total tremor score is even higher (>20) [26]. A study of what was referred to as “borderline ET cases” (i.e., FD-ET with borderline clinical findings who did not meet strict criteria for ET) revealed a mean total tremor score of 11.4 ± 2.6, in comparison with 7.3 ± 1.8 in normal individuals [14, 23]. Hence, by all measures across numerous other studies, our FD-ET would have been classified as normal [14].

We report here that 27% of FD-ET and 17.1% of Co had a spiral axis score that was > 0, meaning that (1) they had some observable tremor while drawing a spiral and (2) that tremor had an identifiable axis. A far larger percentage had an observable tremor while drawing a spiral: 231 (98.3%) of 237 FD-ET and 101 (96.2%) of 105 Co had a spiral score of 0.5 or higher on one or more of their 4 screening spirals. This reflects the presence of some mild degree of enhanced physiological tremor in nearly all enrollees. Prior studies have shown that mild enhanced physiological tremor is observed in > 95% of normal controls during the administration of simple clinical tests such as drawing spirals [27-29].

A limitation is that we only enrolled first-degree, but no second-degree relatives. In the future, enrollment of second-degree relatives would allow us to compare the two groups of ET relatives, as we might expect to find a higher spiral axis score among first-degree relatives. Furthermore, future studies should assess whether patients with other movement disorders and other tremor disorders (e.g., other cerebellar diseases or PD) exhibit the feature we study here.

This study had several strengths. First, to our knowledge, it is the only study to examine this particular clinical feature, the spiral axis score, in first-degree relatives of ET cases and compare it to controls. As such, we have the study design and data to move our examination of this clinical feature from a case-noncase study [2], as has been done, to a family study design, allowing us to test, using another method, whether this feature is disease-linked. Second, all FD-ET were evaluated prospectively using a standardized research protocol. Third, the large sample size (n = 342), allowed for well-powered analyses.

5. Conclusions

In summary, FD-ET had higher spiral axis scores than Co. These data (1) show that such axes are more common in relatives of ET cases than controls, and (2) raise the possibility that the spiral axis may be an early subclinical feature of ET.

• An identifiable axis on spiral drawing is a clinical feature seen in ET cases.

• We compared unaffected ET relatives and controls.

• This feature was present to the greater degree in ET relatives.

• The spiral axis may be an early subclinical feature of ET.

Abbreviations:

CIRS

Cumulative Illness Rating Scale

Co

controls

ET

essential tremor

FD-ET

first-degree relatives of ET cases

PD

Parkinson’s disease