Using myoelectric signal parameters to distinguish between computer workers with and without trapezius myalgia

Complaints of chronic trapezius muscle pain among computer workers have increased in prevalence during the last decade. Currently there is no clear understanding of the pathophysiological mechanisms involved in affected muscles. The major objective of this work was to determine if measurable electrophysiological differences exist between the trapezius muscles in individuals suffering from trapezius myalgia (TM) and occupation-matched pain-free control subjects. Myoelectric signal (MES) data were recorded from the upper trapezius muscle while subjects with and without myalgia performed a standardized series of postural and arm-holding tasks. MES variables reflecting muscle fatigue, muscle tension and motor control strategies were analyzed to determine their potential ability to distinguish between the two groups. One variable, RestTime, was found to be significantly different between the groups but it was not specific enough to predict group association. A multivariate logistic regression analysis yielded a model that separated the two groups with better than 70% sensitivity and 70% specificity. The variables included in the model reflect differences in trapezius muscle activity between the groups, particularly related to motor control and/or active muscle tension, but not fatigue. The model was tested using a small sample of new data, which again produced a good sensitivity (85.7%) but not specificity (42.9%). To the authors' knowledge, this is the first objective MES-based model that has successfully classified subjects with or without TM based on a simple clinical test. Further work with this model might result in understanding the pathophysiology of TM, assisting with clinical diagnosis, and testing the effect of various treatment interventions.