Objective: To explore the long-term variability and reliability of tremor severity in Archimedes spirals in patients with Essential Tremor (ET).

Background: ET affects activities of daily living such as eating, drinking, and writing. Through accelerometry, ET has been shown to vary daily up to 30% in untreated patients. While spiral analysis has been proposed as a valid method to quantify action tremor, data on variability in an at-home setting is scarce. We aimed to explore the daily and weekly variability of spiral tremor and provide data to inform future power analyses in clinical trials.

Methods: 13 ET patients completed hourly tremor diaries including Archimedes spirals during the waking day over a 14-day period. Dominant hand spirals were later scored using the 0-10 Bain and Findley scale in a blinded, randomized fashion, and a subset was rated by two investigators to assess inter-rater reliability using intra-class correlation (ICC) and Kappa. Reproducibility of spiral scores within subjects over time was calculated using ICC. For each subject, four means were calculated (AM and PM at week 1 and week 2), which were then log-transformed. A mixed model was applied to investigate the effect of week (week 1 vs. week 2) and time (AM vs. PM), with both week and time as within-subject factors and time nested within week.

Results: Subjects completed a mean of 145 spirals each, averaging 10.3 (range 5.3-13.6) spirals per day. Among the selected spirals, the inter-rater reliability was high (ICC=0.93, Kappa=0.85). Mean spiral score over the 14-days across subjects was 3.4±0.9 (range of mean 1.3-6.6, SD 0.6-1.5), and the mean coefficient of variation of 28.7 (range 15.5-46.3). The time effect was significant, with patients performing better in the PM than AM (p<0.001), but week effect was not (p=0.096). Reproducibility was high for spiral means on a daily (ICC AM=0.82, ICC PM=0.91) and weekly level (ICC AM=0.90, ICC PM=0.93).

Conclusions: There is considerable variation in spiral severity, both within and across days, with non-random diurnal patterns suggestive of improved and less variable spirals in the afternoons compared to the mornings. The reliability of spirals across days and from week to week is excellent. Our data shows that diary-based capture of tremor spirals may be a feasible and reliable method to continuously monitor spiral tremor severity in an at-home setting, and can be applied in power analyses for future ET trials.

« Back to 2017 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/variability-of-spiral-tremor-during-daily-life-of-subjects-with-essential-tremor/