Does metronome really help timing gait in Parkinson's disease?

Abstract

Unstructured gait variability is a hallmark of timing gait disorders of Parkinson's disease (PD). Recently, complex temporal structure of stride duration variability has been demonstrated in healthy gait pattern, displaying long-range autocorrelations (LRA). Conversely, the breakdown of such temporal organization was associated to dynamic instability in PD. To improve timing gait in PD, synchronization of walking with rhythmic auditory stimulation (RAS) like music or metronome is largely used in clinical settings. However, while structure of RAS can modulate LRA in healthy gait pattern, it remains unknown in PD. Patients performed overground walking trials at a comfortable speed while listening different structures of RAS (counterbalanced order across patients). Each structure was adapted to the gait cadence of each patient as previously measured in a 10-meter walking test. The RAS varied in term of fluctuation of the metronome period in the sound sequence, yielding four conditions: strictly metronomic, randomly fluctuating around the metronome period, fluctuating according to an LRA structure (Hurst exponent = 0.80) and no auditory stimulus. Cadence, speed, stride length, temporal organization (LRA) and magnitude of stride duration variability were measured on 512 consecutive gait cycles. The presence of LRA was based on scaling properties of the series variability (Hurst exponent) and the shape of the power spectral density (α exponent). Those measures were compared across the four conditions using a one-way repeated ANOVA. Our results show that temporal organization of PD gait may be modulated using different auditory structures. Spearman's coefficients indicate adequate correlations between LRA of gait and auditory cue. This matching indicates strong adaptation and synchronization of the gait to the RAS in those patients. However, LRA were systematically lower during auditory conditions compared to spontaneous walking session. Furthermore, magnitude of stride duration variability, walking cadence, walking speed, stride length were not statistically different across different conditions. Consequently, strictly metronomic auditory stimuli do not seem to be an optimal way to improve timing gait in PD, as it induces the disappearance of LRA. Future work will investigate whether structured auditory stimuli induce gait improvement in PD.