Neck Muscle EMG-Force Relationship and Its Reliability During Isometric Contractions

Riccardo Lo Martire, Kristofer Gladh, Anton Westman, Björn O. Äng

Susceptible to injury, the neck is subject to scientific investigations, frequently aiming to elucidate possible injury mechanisms via surface electromyography (EMG) by indirectly estimating cervical loads. Accurate estimation requires that the EMG-force relationship is known and that its measurement error is quantified. Hence, this study examined the relationship between EMG and isometric force amplitude of the anterior neck (AN), the upper posterior neck (UPN), and the lower posterior neck (LPN) and then assessed the relationships' test-retest reliability across force-percentiles within and between days. EMG and force data were sampled from 18 participants conducting randomly ordered muscle contractions at 5-90% of maximal voluntary force during three trials over 2 days. EMG-force relationships were modeled with general linear mixed-effects regression. Overall fitted lines' between-trial discrepancies were evaluated. Finally, the reliability of participants' fitted regression lines was quantified by an intraclass correlation coefficient (ICC) and the standard error of measurement (SEM). A rectilinear model had the best fit for AN while positively oriented quadratic models had the best fit for UPN and LPN, with mean adjusted conditional coefficients of determination and root mean square errors of 0.97-0.98 and 4-5%, respectively. Overall EMG-force relationships displayed a maximum 6% between-trial discrepancy and over 20% of maximal force, and mean ICC was above 0.79 within day and 0.27-0.61 between days across areas. Corresponding SEM was below 12% both within and between days across areas, excluding UPN between days, for which SEM was higher. EMG-force relationships were elucidated for three neck areas, and provided models allow inferences to be drawn from EMG to force on a group level. Reliability of EMG-force relationship models was higher within than between days, but typically acceptable for all but the lowest contraction intensities, and enables adjustment for measurement imprecision in future studies.