Prospective Blinded Comparison of Surface versus Fine-wire Electromyographic Analysis of Muscle Recruitment

Background: The shoulder joint, being the most mobile joint in the body, is often affected by instability. Shoulder instability is a symptomatic, abnormal motion of the glenohumeral joint which can result in pain, subluxation or dislocation. Surgery and physiotherapy both have a place in the management of shoulder instability. However, identifying the cause of the instability is essential to determine the correct treatment and prognosis. The Stanmore classification system acknowledges that shoulder instability can be secondary to multiple pathologies that exist contemporaneously within a shoulder. Patients undergo diagnostic tests including electromyography (EMG) tests to help confirm whether atruamatic instability is due to structural or abnormal muscle patterning. Several studies have used EMG or cadaveric models to reveal that abnormal muscle activation in pectoralis major (PM), anterior deltoid (AD), latissimus dorsi (LD) and infraspinatus (IS) is present with shoulder instability. As the aetiology in muscle patterning cases is essentially muscular, authors suggest that non-operative treatment is considered first. Fine-wire electromyography (fwEMG) is used as a diagnostic tool in shoulder instability to asses muscle patterning and plan intervention. However, this procedure is invasive, and not widely available in clinical settings. Surface electromyography (sEMG) is a non-invasive technique, which is thought to be less reliable in detecting deeper muscle activity although no current evidence exists to support this view. Aims: The primary aim of this study is to evaluate the level of agreement between surface and fine-wire electromyography in participants with no known shoulder pathology to establish whether the two techniques would be comparable in their reliability in detecting abnormal patterns. Secondary objectives include the assessment of the repeatability and inter-rater agreement of fwEMG and sEMG. Hypothesis: It was hypothesized that there will be a difference in the results obtained by bother techniques when applied to normal subjects to evaluate muscle patterning around the shoulder. Additionally, it was hypothesized that there will be a difference in repeated fwEMG and sEMG signals and differences in interpretation of the signals. Methods: Ethical approval was obtained. Subjects were recruited from the Institute of Orthopaedics and Musculoskeletal Science at the Royal National Orthopaedic Hospital. Subjects were included if they were aged between 18 – 60, had no previous or current history of shoulder, neck or back pain, and had a good understanding of English. They were excluded if they were able to voluntarily sublux or dislocate the shoulder, had a fear of needles, or had a BMI > 30. Twelve participants gave informed consent. The EMG system used was the Sapphire II (Medelec Ltd, Surrey, England), with a double-differential detection mode where the electrical signals were transmitted on to a four-channel oscilloscope displaying the raw EMG signal and prints displays. The following settings were used. Table 1: Settings used on the Sapphire II EMG System Electrode Sensitivity Timebase Low Filter High Filter Fine-wire 1.00mV 10s 3Hz 10kHz Surface 200μV 10s 3Hz 10kHz A dual-needle insertion technique was used to insert two separate hypodermic needles containing fine-wire intramuscular electrodes fabricated from Teflon-coated tungsten wire into pectoralis major, anterior deltoid, infraspinatus and latissimus dorsi (PM, AD, IS and LD) at standardised locations. Ambu Surface triple-stud self-adhesive electrodes were attached on the same locations. Participants performed five identical uni-planar standard movements (flexion, abduction, external rotation, horizontal flexion with adduction and extension) while connected to fine-wire electrodes. The same procedure was repeated using surface electrodes. Each of the procedures was repeated to assess repeatability. Two clinicians, experienced in electromyography, interpreted all the traces to report if muscle activation patterns differed from or agreed with the accepted “normal”. Sensitivity, specificity and Kappa values for level of agreement between electrode-type, repeatability and inter-rater agreement were calculated according to the method of Altman (1991). Results: Ten participants (of which 7 are females, 3 males) completed the study between 2008 and 2009. Table 2: Participant characteristics