Nonlinear dynamic analysis of oscillatory repetitive movements in Parkinson's disease and essential tremor

• Published in: Movement disorders Vol. 25; no. 15; pp. 2577 - 2586
• Main Authors: Costa, João, González, Hernán A., Valldeoriola, Francesc, Gaig, Carles, Tolosa, Eduardo, Valls-Solé, Josep
• Format: Journal Article Publication
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 15.11.2010; Wiley; Wiley Subscription Services, Inc
• Subjects: Adult; Aged; Aged, 80 and over; Biological and medical sciences; Degenerative and inherited degenerative diseases of the nervous system. Leukodystrophies. Prion diseases; Enginyeria mecànica; essential tremor; Essential Tremor - physiopathology; Female; Fingers - physiopathology; Forearm - physiopathology; Humans; Male; Mechanical engineering; Medical research; Medical sciences; Middle Aged; Movement disorders; Neurology; neurophysiology; Nonlinear Dynamics; Parkinson Disease - physiopathology; Parkinson's disease; Parkinson, Malaltia de; Psychomotor Performance; signal processing; Signal Processing, Computer-Assisted; Statistics, Nonparametric; Tremor - physiopathology; Àrees temàtiques de la UPC; Nervous system diseases; Parkinson's disease; Tremor; Parkinson disease; Involuntary movement; Cerebral disorder; Neurophysiology; Central nervous system disease; Signal processing; Nonlinearity; essential tremor; Degenerative disease; Neurological disorder; nonlinear dynamics; Extrapyramidal syndrome
• ISSN: 0885-3185; 1531-8257; 1531-8257
• DOI: 10.1002/mds.23334

Uncertainty exists on whether Parkinson's disease (PD) and essential tremor (ET) patients have similar degree of impairment during motor tasks. We investigated this problem by analyzing nonlinear dynamics of repetitive movements in 21 control subjects, 33 mild‐moderate PD patients, and 18 ET patients. Accelerometer signals were recorded during finger tapping and unbounded forearm movements between two points, and processed with moving average filtering to generate a new signal consisting of the temporal distance between consecutive cycles. We calculated: mean interpeak interval (slowness), interpeak interval variability (irregularity), and beat decay (BD) of the auto mutual information (AMI) value, which estimates signal predictability by measuring the loss of signal information over a timescale. Both PD and ET had longer interpeak interval (except for finger tapping), higher interpeak interval variability, and higher BD‐AMI values than controls (P ≤ 0.007, all comparisons). ET patients had higher BD‐AMI values than PD (P = 0.003). BD‐AMI was the parameter that discriminated better between subjects (diagnosis accuracies about 80%). No differences existed between PD patients with and without tremor or between PD or ET patients with different disease stages, for any parameter. Evaluation of nonlinear dynamics of oscillatory repetitive movements is a feasible and promising tool for studying movement physiology. Movement performance is more predictable in PD and ET than in controls, even in early disease stages. Slowness and irregularity of movement in PD and ET cannot be fully explained by tremor. Some common pathogenic mechanisms leading to bradykinesia may contribute to this impairment. © 2010 Movement Disorder Society