One-year reproducibility and stability of the signal amplitude ratio and other variables of the electromyogram: test-retest of a shoulder forward flexion test in female workers with neck and shoulder problems.

We have studied 23 women with neck and shoulder problems in a car factory with a 1-year interval. Our main aim was to investigate the reproducibility and stability of the tension pattern. In addition, the mean frequency (MNF) of the power spectrum and the signal amplitude (RMS) of the surface electromyograph (EMG) of the trapezius, deltoid and infraspinatus muscles and mechanical output were determined throughout 100 maximal isokinetic shoulder forward flexions. The signal amplitude ratio (SAR) was calculated as the ratio between the signal amplitude of the EMG of the passive relaxation and the active flexion part of each contraction cycle. The SAR variable can be reproduced (r = 0.47-0.76) with a 1-year interval. There was a significantly lower SAR of the trapezius at the second test, which might have been due to lower work pace at the factory. The longitudinal patterns of SAR throughout the two tests were similar at both tests. There were significant correlations between tests for 18 out of 22 EMG variables, even though the correlations were generally lower than for SAR (initial MNF: r = 0.39-0.48; MNF endurance level: r = 0.55-0.83; RMS (%): r = 0.08-0.46). Peak torque had better reproducibility than work. In conclusion, SAR has a long-term reproducibility equal to or better than other EMG and biomechanical variables. The present results indicate that SAR has potential to measure unnecessary muscle tension in intervention studies and to identify individual movement patterns.