Reliability of spike and turn variables of surface EMG during isometric voluntary contractions of the biceps brachii muscle.

The present study was performed to investigate the reliability of initial value and the rate of change of eight spike and turn variables in biceps brachii muscle of eleven subjects during sustained isometric voluntary contractions. Three levels of contraction were studied: 20%, 50% and 80% of the maximal voluntary contraction (MVC) level. Reliability was investigated using the intraclass correlation coefficient (ICC) and the standard error of measurement (SEM) of the estimates for each subject. The reliability of the spike and turn variables was compared with four classical variables (ARV: average rectified value, RMS: root mean square MDF: median frequency and MNF: mean frequency) for the same recorded signals. Tukey's HSD (honestly significant difference) test was applied to detect significant differences at the 95% confidence level (p<0.05) between the %MVC levels for each of the eight variables. The main findings of this study can be summarized as follows: (i) the initial value estimation with eight spike and turn variables showed good reliability but the estimation of rate of change showed poor reliability. (ii) The reliability of the classical variables for estimation of the initial value was slightly lower than that of the spike and turn variables, but the four classical variables showed better reliability than the spike and turn variables for estimation of the slope. (iii) The initial values of the amplitude variables (MTA: mean turn amplitude and MSA: mean spike amplitude) increased and the rate of changes of the frequency variables (MTF: mean turn frequency, MSF: mean spike frequency and ZC: zero-crossing) decreased significantly when the contraction level increased from 20% MVC to 80% MVC.