Reliability of sEMG data of back muscles during static submaximal loading situations - Values and pitfalls.

Alterations of trunk muscle activity can be objectified using surface electromyography (sEMG). But for diagnostic purposes reliable reference values are needed. Therefore, the present study evaluated sEMG measurements of representative back muscles concerning their reliability. For this purpose, 49 healthy individuals (24 women) were subjected to static loads with defined portions of their upper body weight by applying whole-body tilts up to 90° in upright posture. Reliability analyses were carried out for amplitude values and normalized values according 90° tilt angle. Reliability measures included intraclass correlation coefficient (ICC), standard error of measurement (SEm), standard error of the mean (SEM), and coefficient of variation of method errors (CVME). ICC values for the amplitude values were > 0.79 for M. multifidus (MF) and M. longissimus (LO). For the normalized values, distinctively lower ICC levels were generally detected (0.33-0.81). The values of SEm and SEM also increased with increasing load, while CVME levels of the amplitude values maintained similar levels. For defined static extension loads, good to excellent reliability can be demonstrated for back muscles. The established normative values for SEm, SEM and CVME parameters may be used for diagnostic purposes during defined submaximal load situations. This does not apply to normalized data.