Perioperative quantification of MDS-UPDRS-III tremor measurements in patients with Parkinson's disease using accelerometry.

Tremor in patients with Parkinson's disease (PD) is usually evaluated using the Movement Disorder Society Unified PD Rating Scale (MDS-UPDRS). Limitations of the MDS-UPDRS include subjectivity and rater-dependency, which may be overcome by accelerometry. This study focused on relating the MDS-UPDRS hand tremor scores to an objective scoring method with 3D-accelerometry. An accelerometric algorithm to measure and classify tremor of the hands according to MDS-UPDRS criteria is proposed. Sixty-four PD patients and 64 healthy controls (matched on sex and age) were included. Two raters assessed MDS-UPDRS tremor-criteria, while accelerometry was performed at the index finger. Measurements were executed in an off-medication state, after a washout period of at least 12 h. The 3D-acceleration data included amplitude and area-under-the-curve of power in the 4-7 Hz range. Agreement between MDS-UPDRS and accelerometric scores was analyzed with Cohen's kappa coefficient (κ). The trends between tremor amplitude and MDS-UPDRS score were consistent with the logarithmic relationship reported in previous studies. Overall, the accelerometric scores showed a substantial agreement (> 80.2%, κ ≥ 0.732) with the tremor amplitude according to MDS-UPDRS ratings. Accelerometric constancy of rest tremor (CRT) measures correlated less with MDS-UPDRS scores (R2 < 0.469, p < 0.001) than the other tasks. Accelerometric test-retest reliability was good to excellent (ICC ≥ 0.789, p < 0.001). MDS-UPDRS tremor tests can be reliably translated to objective accelerometric measurements. However, discrepancies were found between accelerometric CRT-measures and MDS-UPDRS ratings. Accelerometry could complement visual evaluations of tremor, aiming to reduce rater-variability of MDS-UPDRS measurements and allow for more objective monitoring of tremor in the clinical setting.