Adhesive capsulitis, also known as frozen shoulder (FS), is a condition characterized by the functional restriction of both active and passive shoulder motion with unremarkable glenohumeral joint radiographs findings. Shoulder pain almost exists during the whole course of frozen shoulder. Pain induces muscle spasm and causes pain-spasm-pain cycle resulting in limited range of motion, changed muscle recruitment pattern and finally influences neuromuscular control. Commonly used conservative interventions for FS patients include joint mobilization, corticosteroid injection, exercise, modality and soft tissue mobilization. Despite joint mobilization is the most commonly used manual therapy in patients with FS, the evidence level is weak and the efficacy is not superior to other conservative treatments. Soft tissue mobilization is widely used in lots of musculoskeletal conditions. The effects of soft tissue mobilization include breaking the adhesion tissue and improving range of motion, muscle strength and motor control. However, few studies have investigated the effect of soft tissue mobilization in patients with frozen shoulder. Therefore, the purpose of this study is to investigate and compare the effect of soft tissue biased manual therapy and conventional physical therapy in patients with primary FS.

Adhesive capsulitis, also known as frozen shoulder (FS), is a condition characterized by a functional restriction of both active and passive shoulder motion with unremarkable glenohumeral joint radiographs findings. Frozen shoulder is divided into primary and secondary type. The mechanism of primary FS is still unknown and the secondary FS may accompany with shoulder injuries. The incidence of primary FS is 2 to 5.8% and the risk factors are diabetes and thyroid disease.

The course of FS is divided into four consecutive stages which are inflammation, painful, frozen and thawing phase. Shoulder pain almost exists during the whole course of frozen shoulder. The pain and discomfort of FS patients are frequently localized to the deltoid insertion and coracoid process. Pain induces muscle spasm and causes pain-spasm-pain cycle resulting in limited range of motion, changed muscle recruitment pattern and finally influences neuromuscular control.

Commonly used conservative interventions for FS patients include joint mobilization, corticosteroid injection, exercise, modality and soft tissue mobilization. Despite joint mobilization is the most commonly used manual therapy in patients with FS, the evidence level is weak and the efficacy is not superior to other conservative treatment s. Soft tissue mobilization is widely used in lots of musculoskeletal conditions. The effects of soft tissue mobilization include breaking the adhesion tissue and improving range of motion, muscle strength and motor control. However, few studies have investigated the effect of soft tissue mobilization in patients with frozen shoulder. Only one study used one-time muscle release to FS patients and investigate improvement in shoulder biomechanics, muscle strength and ROM. Therefore, the purpose of this study is to investigate and compare the effect of soft tissue biased manual therapy and conventional physical therapy for six weeks in patients with primary FS.

The study design is a pretest-posttest control group design. The investigators plan to recruit 70 patients with primary frozen shoulder and divide them into a soft tissue biased manual therapy group and a conventional physical therapy group. The sample size is determined by the previous study. It is based on a significance level of 0.05, and a power of 0.80. The outcome measures in this study include scapula kinematics, scapula position, the range of motion, muscle tone, muscle strength, pain and functional impairment.

LIBERTY™ electromagnetic tracking system (Polhemus, Colchester, VT, USA) was used to collect three-dimensional kinematic (3D) data during scaption, hand to neck and hand to back tasks at a sampling rate of 120 Hz, and the software Motion Monitor® (Innovative Sport Training, Inc., Chicago. IL. USA) was used to analyze the data. The main measurements of shoulder kinematics include scapular upward/downward rotation, internal/external rotation, and anterior/posterior tilt. A stylus was used to digitize the bony landmarks for defining the anatomical coordinate system. The methods for this measure have been described previously.

For collecting muscle activation data, the investigators used surface electromyography (sEMG, TeleMyo 2400 G2 Telemetry; Noraxon USA, Inc., USA) to collect scapular muscles' activation during those tasks. The investigators will measure the muscle activities of the pectoralis major, upper trapezius, infraspinatus, teres major and lower trapezius. The electrodes will be placed according to previous studies and be positioned in parallel to the direction of the muscle fibers.

The range of motion of shoulder flexion, abduction, external rotation and internal rotation will be measured by plastic universal goniometer according to Norkin's methods. The muscle strength of lower trapezius, external rotators and internal rotators are measured by the hand-held dynamometer. The scapula positions are measured by modified scapular sliding test with calliper during arms by sides, hands on the hips and scaption 90 degrees. The muscle tone of pectoralis major, upper trapezius, infraspinatus and teres complex will be measured by hand-held myotonometer (Myoton-Pro, Myoton AS, Tallinn, Estonia) in shoulder resting position according to previous studies. The functional impairments are measured by Flexilevel Scale of Shoulder Function questionnaire. The pain level is measured by the visual analogue scale.

Participants in the soft tissue biased manual therapy group will receive heat and muscle release for six weeks and participants in conventional physical therapy group will receive modalities and joint mobilization for six weeks. The measurement will be obtained at baseline, 3 weeks, and after the intervention. Repeated measures ANOVA will be used for data analysis. The level of significance sets at α=0.05.

Inclusion Criteria:

• Primary adhesive capsulitis diagnosed by orthopaedic surgeon or physiatrist
• Reduction in range of motion in more than two planes
• < 30 degrees of the external rotation range of motion during shoulder abduct to 90 degree
• No radiologic abnormality
• Onset time more than three months
• No night pain and resting pain

Exclusion Criteria:

• History of stroke, diabetes mellitus, rheumatoid arthritis, rotator cuff tear, shoulder osteoarthritis, surgical stabilization/surgery of the shoulder, osteoporosis, or malignancies in the shoulder region