Background: The Archimedes spiral test is widely used in clinical neurology to evaluate tremor, yet its diagnostic performance for essential tremor (ET) remains unclear and methodologically inconsistent across studies. Objective: To determine the diagnostic accuracy of the Archimedes spiral test for distinguishing ET from healthy controls (HC) and to identify methodological and technological factors associated with improved performance. Methods: A systematic review and meta-analysis were conducted according to PRISMA 2020 guidelines (PROSPERO: CRD420251167793). Eight studies (1,046 participants) were included. Risk of bias was assessed with QUADAS-2. Pooled diagnostic accuracy was estimated using a random-effects model, and subgroup/meta-regression analyses examined the impact of digital acquisition, AI algorithms, EMG/accelerometry, and execution parameters. Results: The pooled diagnostic accuracy was 0.89 (95% CI: 0.77–0.95; I² = 75.4%). Technology-assisted methods (digital capture or AI-based analysis) demonstrated higher accuracy (0.91; 95% CI: 0.79–0.96) than manual approaches (0.78; 95% CI: 0.70–0.84), although differences did not reach statistical significance. Exploratory analyses suggested that medication withdrawal, absence of arm support, digital recording, and standardized execution protocols may contribute to improved diagnostic performance. Conclusions: The Archimedes spiral shows moderate-to-high diagnostic accuracy for ET, particularly when combined with digital acquisition and objective analytic methods. Given the small number of available studies and substantial methodological heterogeneity, these findings should be interpreted as exploratory. Standardized digital protocols may enhance reproducibility and support the development of scalable, clinic-ready tremor biomarkers.
Macedo et al. (Fri,) studied this question.