Dynamic sEMG analysis of gait in Parkinson’s patients with freezing in ON

Abstract

Introduction: Freezing of gait (FOG) represents one of the most disabling and least understood symptoms in PD, strongly associated with falls. Episodes of FOG are characterized by a sudden but transient motor block occurring at the beginning or during walking performance. The kinematics analysis of gait pointed out that the steps of PD patients with FOG present, during their ‘normal’ walk, alterations that are common for all PD subjects: reduced velocity and step width, altered anticipatory corrections both laterally and antero-posteriorly, but is unclear the connection with FOG. Recently it was hypothesized that the stride to stride variability of gait could contribute to this phenomenon with patterns of premature timing of distal muscles before freezing [1,2,3]. The aim of this study was to quantify the electromyographic activity of the distal and proximal muscles during three steps just before freezing, in order to accurately analyse the alterations in the muscular activation strategies and to evaluate hypotheses about the ON freezing generation in PD. Materials and methods: 10 Patients (age 74, SD � 8.4) with PD and FOG (UDRS 15.4—FOGQ 14.3) were included. Surface EMG data were collected bilateral by using a portable commercial wireless system (PortiLab2) that acquires data coming from tibialis anterior (TA), gastrocnemius (GS), gluteus maximus (GX) and gluteus medius (GM) muscles together with six basographic (footswiches) and three accelerometric (3D sensor) signals. Subjects were instructed to walk, in the On phase of the medication cycle, in normal and freezing provoking circumstances. SEMG signals was filtered and then rectified and smoothed in order to obtain the sEMG envelopes with an ad hoc software written in MATLAB. The sEMG signals are then synchronized with gait phases and the characteristic of the patterns during gait just before freezing are visualized and analysed. Results: The data show significantly abnormal timing of TA (p < 0.005) and GS with overall preserved reciprocity. Just before freezing onset TA and GS swing activities prematurely started in the pre swing phase. Similar timing alterations are present in all the examined muscles activities (GM and GX) with an inversion of activation pattern between omo and controlateral muscles in the swing phase. The altered activation mainly present on the side more compromised by Parkinson’s disease. Discussion: The altered pattern of distal and proximal muscles activation in the swing phase could compromise, before freezing, gait forward progression. These findings could support the hypothesis that the freezing in ON is not a sudden phenomenon, as it could seem to the clinical observation, but it is rather compatible to a progressive disorganization in the gait planning, that can be initially compensated by an adaptation of the motor control. This altered pattern could be associated to the more compromised side by the PD. The quantitative analysis could permit a more specific decision to the best rehabilitative intervention to be carried out with these