[ARTICLE] Brain oscillatory activity as a biomarker of motor recovery in chronic stroke – Full Text

Abstract

In the present work, we investigated the relationship of oscillatory sensorimotor brain activity to motor recovery. The neurophysiological data of 30 chronic stroke patients with severe upper‐limb paralysis are the basis of the observational study presented here. These patients underwent an intervention including movement training based on combined brain–machine interfaces and physiotherapy of several weeks recorded in a double‐blinded randomized clinical trial. We analyzed the alpha oscillations over the motor cortex of 22 of these patients employing multilevel linear predictive modeling. We identified a significant correlation between the evolution of the alpha desynchronization during rehabilitative intervention and clinical improvement. Moreover, we observed that the initial alpha desynchronization conditions its modulation during intervention: Patients showing a strong alpha desynchronization at the beginning of the training improved if they increased their alpha desynchronization. Patients showing a small alpha desynchronization at initial training stages improved if they decreased it further on both hemispheres. In all patients, a progressive shift of desynchronization toward the ipsilesional hemisphere correlates significantly with clinical improvement regardless of lesion location. The results indicate that initial alpha desynchronization might be key for stratification of patients undergoing BMI interventions and that its interhemispheric balance plays an important role in motor recovery.

1 INTRODUCTION

Stroke is a major global health problem. The number of stroke victims has been rising in the past years all around the world. Millions of stroke survivors are left with very limited motor function or complete paralysis and depend on assistance (Feigin et al., 2016). Therapeutic approaches such as constraint‐induced movement therapy are not applicable to the group of patients with severe limb weakness (Birbaumer, Ramos‐Murguialday, & Cohen, 2008). However, brain–machine interface (BMI) training has demonstrated efficacy in promoting motor recovery in chronic paralyzed stroke patients (Ramos‐Murguialday et al., 2013), and long term effects (Ramos‐Murguialday et al., 2019). Subsequent work has replicated and confirmed BMI efficacy. Arm and hand movements are trained using a body actuator (e.g., orthotic robots) that is controlled by oscillatory activity of the brain (Ang et al., 2014; Frolov et al., 2017; Kim, Kim, & Lee, 2016; Leeb et al., 2016; Mokienko et al., 2016; Ono et al., 2014). Brain signals can thus travel to the limb muscles along an alternative pathway. Contingently linking movement‐related patterns of brain activity and visuo‐proprioceptive feedback of the movement supports associative learning (Ramos‐Murguialday et al., 2012; Sirigu et al., 1995).

Changes in sensorimotor brain oscillations involved in planning and execution of movements were used as control signals for the BMI in the aforementioned studies. The sensorimotor rhythm (SMR) is an oscillation within the alpha frequency range of the EEG during a motionless resting state over the central‐parietal brain regions. Movement planning, imagination and execution lead to its suppression. In the present work, we investigate EEG brain oscillations of the alpha frequency, ranging from 8 to 12 Hz, over the motor cortex, and we term them “alpha oscillations.”

Biomarkers could be defined as indicators “of disease state that can be used as a measure of underlying molecular/cellular processes that may be difficult to measure directly in humans” (Boyd et al., 2017). When dealing with a condition as heterogeneous as stroke validated biomarkers of recovery could help plan treatments and support efficient allocation of resource while maximizing outcome for the patients. Alpha brain oscillations have been evaluated as markers of ischaemia and predictors of clinical outcome in acute patients (Finnigan & van Putten, 2013; Rabiller, He, Nishijima, Wong, & Liu, 2015). Desynchronization in the alpha frequency range has also been investigated as a marker of stroke and a predictor of recovery in the same patient group. Tangwiriyasakul, Verhagen, Rutten, and Putten (2014) showed that the recovery of motor function was accompanied by an increase of alpha desynchronization on the ipsilesional side. In subacute patients presenting mild to moderate motor deficits recovery lead to a similar increase of alpha desynchronization on the affected hemisphere (Platz, Kim, Engel, Kieselbach, & Mauritz, 2002). Furthermore, first attempts investigated correlations of alpha desynchronization with motor improvements in chronically impaired patients (Kaiser et al., 2012). In a controlled study, a group of subacute patients with severe deficits used motor imagery, guided by a brain–computer interface, in addition to their regular physiotherapeutic rehabilitation protocol. They showed a higher probability for motor improvements with increased alpha desynchronization (Pichiorri et al., 2015).

In the present work, we investigated what changes in the oscillatory activity of the brain a proprioceptive BMI coupled with physiotherapy produces over the course of a training intervention and if these correlate with recovery in severely paralyzed chronic stroke patients. We hypothesized that functional motor improvements are accompanied by an ipsilesional increase and a contralesional decrease in alpha desynchronization indicating reorganization of compensatory brain activity from the contralesional to the ipsilesional hemisphere. We intend to establish alpha oscillatory activity as a biomarker of motor impairment and as a building block of statistical models of stroke neurorehabilitation.[…]

 

Continue —->  Brain oscillatory activity as a biomarker of motor recovery in chronic stroke – Ray – – Human Brain Mapping – Wiley Online Library

image

Figure 1
Schematics of the data acquisition phase and the offline analysis for EEG and EMG. Neurophysiological data was acquired using a 16 channel EEG cap and 4 bipolar EMG electrodes on each arm. EEG data were cleaned from eye movement artifacts and trials containing other artifacts (e.g., cranial EMG, head movements, and so on). EMG data were analyzed to detect compensatory muscle contractions on the healthy upper limb and on the paretic side during resting intervals to identify these trials as contaminated because the muscle activity is a sign of undesired EEG activity. Only data free of artifacts were used for the final analysis of EEG oscillatory activity

, , , , , , , , , , ,

  1. Leave a comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: