The teres major and pectoralis major are adductors and may play a compensatory role for deficit in rotator cuff function. The investigators aim to (1) determine the effects of neuromuscular electrical stimulation (NMES) on the two adductors for the outcomes in symptomatic rotator cuff tear subjects, and (2) evaluate the immediate effect of NMES on the co-contraction of the adductors.Thirty symptomatic rotator cuff tear subjects will have NMES on the two adductors. The investigators will measure the acromiohumeral distance by ultrasonography and scapular kinematics during arm elevation with a three-dimensional motion tracking system. Co-contractions will be calculated by surface electromyography.

Participants Based on the pilot study, a sample size of 30 patients is calculated to provide 80% power with an effect size of 0.5 (acromiohumeral distance difference = 0.9 mm with and without NMES). Rotator cuff tear subjects will be recruited from the Department of Orthopedic surgery or Department of Physical Medicine and Rehabilitation at National Taiwan University Hospital (NTUH). All subjects will be provided written informed consent before inclusion in the experiment. The study is approved by the NTUH Institutional Review Board.

Instrumentation Three-dimensional electromagnetic motion tracking system The Polhemus 3Space FASTRAK® (Polhemus Inc., Colchester, VT, US), an electromagnetic motion analysis system with MotionMonitor software (Innovative Sports Training, Inc., Chicago, IL, US), will be used to track and collect three-dimensional scapular kinematics. Three sensors are attached to the flat bony surface of the acromion and the sternum with adhesive tape, and to the distal humerus between the lateral and medial epicondyles via Velcro straps with adhesive tape (Figure 1). The other sensors are attached by a stylus to digitalize the 3-dimensional positions on bony landmarks including the sternal notch, xiphoid process, C7 vertebrae, T8 vertebrae, T12 vertebrae, anterior and posterior glenohumeral joint, acromioclavicular joint, root of the spine of the scapula, inferior angle of the scapula, and lateral and medial epicondyles to build the shoulder complex model. Each receiver will be sampled at the rate of 120 updates/second divided by the number of sensors.

Ultrasonography (USG) measurements The BenQ T3300 Tablet Ultrasound System (BenQ Medical Technology Corp., Taipei, Taiwan) with a L154BH linear array 4-15MHz broadband transducer will be used for non-invasive evaluation of AHD in 3 actively abducted positions (at 0°, 60°, and 90° of shoulder abduction). Real-time imaging will be performed with rapid B-mode scanning to generate images with the following settings: gain, 66-68; dynamic range, 75; persistence, 3; QScan, 4, for sharpness and smoothing; gray map, 3; Chroma 0; steer angle, 0; and transducer, 5.5 MHz. The minimum measurement distance of this USG is 1 mm with error less than ± 5%.

Neuromuscular Electrical Stimulation (NMES) The GM300TE COMBO NMES (Gemore Technology Co., Ltd., New Taipei City, Taiwan), a portable machine with dual controllable channels and a manual switch, will be applied to the PM and TM. According to a previous study, muscle force generation is modulated by stimulated frequency, pulse width, and intensity. It linearly increases with stimulated amplitude and pulse duration. The electronics of the unit creates the electrical impulses with an output frequency of 50 Hz and a pulse width of 300 microseconds. The synchronous (S) mode and asymmetrical rectangular wave-pulse are used throughout the experiment. The amplitude is set as muscle motor-level contraction with self-perception of maximal tolerable intensity. The time of tetanic stimulation is 7 seconds with 20-second intervals between stimulations to avoid muscle fatigue. The two circular electrodes are used on each adductor to minimize the stimulated interference with the nearby muscles.

Surface electromyography (sEMG) Muscle activation will be collected at 1000 Hz/channel with the resolution of a 16-bit A/D converter to 16-bit D/A analog converter (Model MP 150, Biopac Systems Inc., Aero Camino Goleta, CA, USA). The system includes pairs of silver chloride circular electrodes (diameter: 10 mm; Ludlow Company LP, Chicopee, MA) with an interelectrode distance of 20 mm (center to center). The collected signals are amplified with a Grass AC/DC amplifier (Model 15A12, Astro-Med Inc., West Warwick, RI, USA) with gain of 1000 dB, a common-mode rejection ratio (CMRR) of 86 dB at 60 Hz, and a bandwidth (-3 dB) of 10 to 1000 Hz. Full bandwidth data (captured by the data acquisition and analysis systems software, AcqKnowledge, Biopac, Aero Camino Goleta, CA) are reduced using a root mean square (RMS) algorithm to produce envelopes with an effective sampling rate of 50 samples per second. The mean of the three trials for each task are calculated for data analysis.

Force measurement system The force-sensitive measurement system (FlexiForce ELFTM, New Taipei City, Taiwan) combines three single-point FlexiForce B201 sensors, one handle containing USB-interface electronics, and Windows-compatible software. Three circular sensors (diameter 9.53 mm; thickness 0.203 mm) are able to detect the range of force as low (4.4-111 N), medium (111-667 N) and high (667-4448 N) levels, respectively. This ensures that the various forces during maximal voluntary isometric contraction can be measured by the appropriate sensor. When the sensor detects the force, the software displays a histogram, curve graph, or number of the force detected as the real-time bio-feedback. The sampling rate of the data collection is set at 200 Hz.

Procedures Male subjects are asked to remove their shirts, and female subjects are asked to wear an open-backed bathing suit throughout the experiment. The subjects are asked to do the isometric abducted and adducted maximal isometric contraction test against the force sensors on the customized iron-ring, respectively. The maximum force is simultaneously recorded as the mean value during the isometric maximal isometric contraction tasks. Then acromiohumeral distance is assessed by USG probe in 3 abducted positions (resting position, 60° and 90°) with a custom-designed dynamic goniometer surrounding the distal part of the humerus. The USG probe is applied parallelly to the humerus shaft and over the anterior aspect of the anterior acromial margin. Three images are required for each position. Twenty-second intervals between each image prevents muscle fatigue. Following the USD imaging, sEMG electrodes are placed on the teres major and pectoralis major after the area is wiped clean with alcohol and scrubbed to reduce the impedance of the skin. After placement of the electrodes, the subject hang his/her arm in a neutral sitting posture with a force measurement sensor contacting the wrist. Each subject performs five trials of the isometric abduction and adduction task respectively at 60% of maximal force based on real-time force value feedback.

After that, the scapular kinematics during arm elevation are measured by FASTRAK in 5 trials. First, a physical therapist attachs the sensors to build the local coordinate system for the shoulder. Then the operator asks the subject to elevate the arm to ensure the angle was over 90° and to retract the scapula to confirm the axial direction of scapular internal/external rotation displayed on the system. Second, the subjects are instructed to elevate the arm as high as possible at a comfortable speed within 3 seconds and then slowly to lower the arm to the starting position. An audible metronome set to 1 Hz is used as the cue during the elevation task.

All subjects are asked to perform the same assessment with NMES. The order of NMES application to the teres major or pectoralis major will be randomized. All the optimal NMES attachments are confirmed with the recruiting activity by the motor point pen. Subsequently, the maximal tolerable intensity is ensured and recorded in the three abducted positions. Then the scapular kinematics during arm elevation and the acromiohumeral distance measurements in specific arm positions are performed during NMES on one of the target adductors. Subjects are asked to hold the specific arm position with NMES for at least 3 seconds in each trial during the acromiohumeral distance measurement. In addition, the NMES is sustained for at least 7 seconds during the scapular kinematic tracking during arm elevation. After a 10-minute break following the assessments, NMES on another adductor muscle is conducted. Following the NMES stimulation, subjects performs the co-contraction assessment immediately. The muscle activity data in both isometric abduction and adduction are immediately collected pre- and post-NMES. Then the data of the scapular kinematics during arm elevation and the acromiohumeral distance at resting position and at 60° and 90° of shoulder abduction are analyzed.

Statistical analysis Data will be analyzed in Statistical Package for the Social Sciences (SPSS), version 22 (International Business Machines Corp., New York, USA). The Shapiro-Wilk test is used to test for normality. With normal distribution, two-way ANOVAs (3 conditions x 3 positions for acromiohumeral distance measurements and 3 conditions x 6 positions of scapular kinematics) will be used. The alpha level is set at 0.05 with two-tailed testing. Without normal distribution, a nonparametric test (Friedman test) is used for data analysis. Co-contraction ratios between baseline and immediately post-NMES are compared by t test or Wilcoxon signed-rank test whether or not distribution is normal.

Inclusion Criteria:

1. between 20 to 85 years old
2. diagnosed with symptomatic rotator cuff tear based on ultrasonography or magnetic resonance imaging
3. pain or discomfort specific to the shoulder.

Exclusion Criteria:

1. a diagnosis of frozen shoulder or adhesion capsulitis
2. glenohumeral joint instability
3. a history of affected shoulder surgery or traumatic dislocation/fracture over the past year
4. symptoms related to the cervical spine in recent 6 months
5. neuromuscular diseases, such as polymyositis and myotonic dystrophy
6. limited arm elevation less than 90°
7. contraindications for NMES.