Design and setting
This trial was a single-blind parallel-group randomized clinical trial in which patients with FS were recruited from physical medicine and rehabilitation outpatient clinics (Imam Reza and Shohada) belong to Tabriz University of medical sciences, Iran between February 2019 and January 2020. The investigation has been done according to the principles of the Declaration of Helsinki and Good Clinical Practice. Patients were provided written informed consent before any measurement was done. The CONSORT guidelines were conformed, and the CONSORT flowchart was prepared to display the current of the patients at every stage of the trial.
Participants
The inclusion criteria were: 1) a diagnosis of unilateral primary or idiopathic FS [30], 2) pain and restricted movement present for more than one month [31] , 3) normal shoulder x-rays findings, 4) normal neurologic examination, and 5) 18 to 60 years of age. Patients with neurological disorders, such as in stroke, as well as those with uncontrolled diabetes, rheumatic disease in the shoulder, malignant tumors, a history of the fracture or subluxation in the affected shoulder, history of surgery to the injured shoulder, recent drug or alcohol usage or addiction, history of corticosteroid injection in glenohumeral joint, glenohumeral joint instability or radiating pain from a cervical severe radiculopathy were excluded. Patients referred to outpatient clinics were recruited by a trained physiatrist via non-random convenience sampling method.
The demographic and clinical characteristics of the patients were measured at the baseline.
Randomization procedures
Patients were randomly allocated, in a 1:1 ratio, to receive one of the subsequent trial treatments: 1) Group 1: The ACJ mobilization + standard physical therapy + exercises and 2) Group 2: Standard physical therapy+ exercises. Concealed allocation was directed by means of computer-generated random numbers generated by a person not included with enrolling patients. The randomized allocation, enrolling patients and assigning them to treatments, was performed by a trained physiotherapist at the outpatient clinic. Patients were requested not to discuss the specific treatment received with the examiner researcher. Research coordinators who evaluated patient end points and data analyst were blinded to the allocated intervention groups through the trial and while executing analyses; however, it was not probable to blind participants or treating therapists.
Treatment Programs
1) Group 1: ACJ mobilization + standard physical therapy + exercises
ACJ mobilization was performed in up to six 30-minute individual sessions within a 3-week period by a single trained therapist. The treatment techniques were anterior to posterior passive accessory glides of the distal end of the clavicle categorized from Grade I–IV [32].The patient is placed in the supine position and the upper limb is placed in a physiological position with the patient's arm clinging to the body and the hand on the abdomen, which causes the capsule to stretch less and the technique to be less painful. The therapist placed the tips of his both thumbs on the anterior surface of the clavicle adjacent to the ACJ and spread his other fingers out for stability and his forearm was situated in line with the posterior movement at the ACJ. Grade I means an intensity of slight amplitude that is used at the start of the ACJ joint ROM, with no pressure on connective tissue. Grade II means an intensity of a little larger amplitude used from the start of the joint ROM to the middle. Grade I and II are suitable for relieving pain. Grade III is used at a large amplitude from the middle of the joint ROM to the start of the constraint. Grade IV is used against the tissue resistance at a slight amplitude to the limited part of the joint. Grade III and IV are applied for stimulating a stretching to relieve joint stiffness in a shorter tissue [33-35].
2) Group 2: Standard physical therapy + exercises
Standard physical therapy was performed in 10 sessions (3 sessions per week). Physical therapy treatments included:
Hot pack (HP) as a surface heat modality on the upper trapezius muscle for 20 minutes at a temperature of 74.5 ° C to reduce muscle spasm
1 MHz ultrasound at an intensity of 1,5 -1,25 W/cm2 for 5 min for shoulder joint capsule and rotator cuff muscle tendon [36].
TENS (Transcutaneous Electrical Nerve Stimulation) in the conventional mode for 25 minutes (low-intensity, small pulse width: 50-200 μs with a frequency of 50-100 Hz) [37].
In both groups, shoulder exercises were performed at home for 15 minutes per time (10 repetitions of each), twice a day with direction by the physiotherapist once per week. These exercises included active ROM exercise (such as pendulum stretch (passively), placing one hand on the back of the neck and one hand on the back of the waist and then vice versa, active range of motion of the scapula, elbow circle, finger walk on the wall), towel stretch behind the back to improve internal and external rotations, stretching anterior capsule whereas standing next to the door and walking pass the door, cross body stretch for the posterior capsule, stretch arm above the head with the assistance of the healthy arm in lying or standing position and isometric shoulder internal and external rotation proceeded into active resistive to the extent that the patient's pain allows. Patients were instructed about how to perform the exercises in a pain-free manner so as to maximize normal muscle function and movement pattern [38] and provided as a pamphlet comprising the image and the description for all the exercises to confirm compliance. Furthermore, patients were instructed to avoid painful activities including the shoulder and were counseled how to apply pain-free approaches to do daily activities.
Primary and secondary outcome measures
The primary outcomes were self-reported shoulder pain-related disability measured by the Shoulder Pain and Disability Index (SPADI) questionnaire [39] and goniometric assessment of active shoulder ROM [33, 40].
SPADI was established to quantity present shoulder pain and disability in an outpatient setting. The SPADI comprises 13 pieces that evaluate two areas; a 5-piece subscale that measures pain and an 8-piece subscale that measures disability [41], each element is recorded on a visual analogue scale (VAS). The patient was taught to place a spot on the VAS for each piece that best signified their experience of the shoulder condition over the last week [42]. Each subscale is summed and converted to a total out of 100. A mean is obtained from the two subscales to give a total score out of 100, greater score representing higher disability [20]. We applied the Persian version of SPADI with confirmed validity and reliability in patients with shoulder conditions in research setting and clinical practice [43].
A metal goniometer was used to measure the ROM of the shoulder joint actively and passively in flexion, abduction and internal and external rotation movements. The ROM was measured in a sitting position while holding the patient's scapula with a stable hand [44].
The Shoulder hand behind back (HBB) ROM method was also applied for measuring active shoulder internal rotation with established validity in subjects with shoulder pain of mechanical origin. HBB ROM was defined by determining the interval in centimeters between the T1 spinous process and the radial styloid process and compared to the healthy side, the greater the difference between the two sides, the worse the ROM. HBB ROM is precise, has respectable inter- and intra-rater reliability, and provides normal values for between-limb ROM variability [20, 45].
The secondary outcome was the Visual Analogue Scale (VAS). VAS was applied to measure the pain intensity. VAS is a fast (statistically quantifiable and reproducible) measurement of pain intensity. Patients were asked to place a vertical mark along a 10-cm-long horizontal line indicating the severity of their pain (0 = no pain, 10 = the worst pain) [21].
Statistical analysis
Using G-POWER 3.1.9.2 (Faul, Erdfelder, Lang, & Buchner, 2007) software and based on the Çelik et al study [25] with 80% power, an alpha level of 0.05 and Effect size =0.44, the total sample size was calculated to be 50 (25 in each group). An allowance was made for a 10% dropout rate, increasing the sample size to 56 patients (28 per group).
All data achieved from the trial were analyzed by the SPSS 17.0 software (IBM SPSS Statistics for Windows, Version 17.0. Armonk, NY: IBM Corp) with the level of significance considered at p<0.05. The normal distribution of different variables was evaluated using the Shapiro-Wilk test. The Mann-Whitney U-test and the χ2 test were applied for intergroup comparisons of baseline data. Mixed ANOVA test (time [within subject] * group [between subjects]) in conjunction with the sidak posthoc as adjustment method were used to estimate within group alterations and between group differences. We termed effect size as Cohen’s d for outcome measures. We measured the effect sized of 0.2, 0.5 and 0.8 as small, medium and large, respectively [43].