Two experiments compared multiple methods of estimating postural stability entropy to address: 1) if postural complexity differences exist between concussed and healthy athletes immediately following return-to-play; 2) which methods best detect such differences; and 3) what is an appropriate interpretation of such differences. First, center of pressure (COP) data were collected from six concussed athletes over the six weeks immediately following their concussion and from 24 healthy athletes. Second, 25 healthy non-athletes performed four quiet standing tasks: normal, co-contracting their lower extremity muscles, performing a cognitive arithmetic task, and voluntarily manipulating their sway. Postural complexity was calculated using approximate, sample, multi-variate sample, and multi-variate composite multi-scale (MV-CompMSE) entropy methods for both high-pass filtered and low-pass filtered COP data. MV-CompMSE of the high-pass filtered COP signal identified the most consistent differences between groups, with concussed athletes exhibiting less complexity over the high frequency COP time-series. Among healthy non-athletes, high-pass filtered MV-CompMSE increased only in the co-contraction condition, suggesting the decrease in high frequency MV-CompMSE found in concussed athletes may be due to more relaxed muscles or less complex muscle contractions. This decrease in entropy may associate with reported increases in intra-cortical inhibition. Furthermore, a single-case study suggested high frequency MV-CompMSE may be a useful clinical tool for concussion management.
Keywords: Brain injury; Concussion; Entropy; Postural stability; mTBI.
Copyright © 2016 Elsevier B.V. All rights reserved.