2.1 Study Design and setting

This was a prospective, single-center RCT conducted over 24 months (March 2024 to February 2025) at the Shohada Specialized Hospital in Tabriz, Iran. The study protocol received approval from the Institutional Ethics Committee of Tabriz University of Medical Sciences (Code: IR.TBZMED.REC.1403.021) and was registered in the Iranian Registry of Clinical Trials (IRCT20240508061712N1). The study was conducted in accordance with the CONSORT guidelines.

2.2 Participants

A total of 186 patients with a diagnosis of full-thickness RCT confirmed by MRI were initially screened. After applying inclusion and exclusion criteria, 120 patients were enrolled and randomly allocated into two groups. Inclusion criteria were: age 40–70 years, diagnosis of full-thickness RCT, failure of at least 3 months of conservative treatment, suitability for general anesthesia, and willingness to provide informed consent and attend follow-up visits. Exclusion criteria included: previous surgery on the affected shoulder, severe glenohumeral arthropathy (Hamada grade ≥3), tears with advanced fatty infiltration (Goutallier grade ≥4 on pre-operative MRI, as this population often has a different prognosis and surgical indication (28)), advanced neuromuscular diseases, uncontrolled diabetes (HbA1c >8 %), smoking >10 cigarettes/day, active rheumatic diseases, and unwillingness to continue participation.

2.3 Data collection: Expanded variables

In addition to the core outcomes, the following data were systematically collected:•Demographic & Baseline: Age, gender, Body Mass Index (BMI), occupation (sedentary/light/heavy manual), smoking history (pack-years).•Clinical & Tear Characteristics: Duration of symptoms (months). Etiology (degenerative vs. acute traumatic). Tear size (in cm, measured on pre-operative MRI sagittal and coronal planes by a musculoskeletal radiologist). Number of tendons involved. Presence and grade of fatty infiltration of the supraspinatus and infraspinatus muscles on pre-operative MRI using the Goutallier classification (Grades 0–4) (29). Pre-operative pain level (VAS 0–10).•Comorbidities: Documented history of Diabetes Mellitus, rheumatoid/autoimmune diseases, and thyroid disorders.•Surgical Data: Procedure type (Arthroscopic/Open). Repair technique (Arthroscopic: Single-Row; Open: Double-Row Suture Bridge). Number of suture anchors used. Quality of tendon tissue intraoperatively (recorded by surgeon as: Excellent/Good/Fair/Poor). Operative time (skin-to-skin, in minutes). Surgeon experience (>10 years, >2000 similar procedures).

Postoperative & Re-tear Data: Length of hospital stay (days). Time to return to normal activities/work. All suspected re-tears were evaluated with a standardized protocol: 1) Clinical assessment for sudden loss of function or new trauma, 2) Repeat MRI at 6 months post-op (or earlier if clinically indicated), 3) Review by a musculoskeletal radiologist blinded to group assignment, 4) Final adjudication by the operating surgeon regarding probable cause (e.g., new trauma, biological failure, technical issue) based on MRI and clinical notes.

2.4 Randomization and interventions

Participants were randomly assigned to either the Early Motion Group (EMG) or the Delayed Motion Group (DMG) using block randomization (block size of 4) via Random Allocation Software (version 2.1). Allocation was concealed in sequentially numbered, sealed, opaque envelopes opened after surgery by an independent staff member.

All surgical procedures were performed by a single senior orthopedic surgeon with over 10 years of experience. Arthroscopic repairs utilized a single-row technique, while open repairs utilized a double-row suture bridge technique.•EMG Protocol: Within 24 h postoperatively, supervised passive forward flexion and external rotation (abduction was not permitted in the first 6 weeks) were initiated. These exercises were performed solely by the physiotherapist with the patient in a supine position. The patient was instructed to perform active wrist, finger, and elbow exercises independently. Active shoulder motion and strengthening were introduced progressively starting at week 6.•DMG Protocol: The shoulder was placed in a sling for strict immobilization for the first 6 weeks. Patients were allowed only active wrist, finger, and elbow exercises. No pendulum or passive shoulder motions were permitted. An identical supervised rehabilitation protocol (as described for the EMG from week 6) was then initiated from week 6 onwards.

2.5 Outcome measures

Primary outcomes at 6 months postoperatively included:1.Pain: Measured using the Visual Analog Scale (VAS; 0–10).2.Function: Assessed via the UCLA Shoulder Score.3.Range of Motion (ROM): Measured with a standard goniometer for forward flexion, abduction, internal rotation, and external rotation.4.Structural Integrity: Evaluated by MRI using the Sugaya classification (Grade I-V; Grades IV & V defined as re-tear).

Secondary outcomes included patient satisfaction, incidence of complications (e.g., stiffness, infection), time to return to daily activities, and mid-term outcomes at 3 months. Assessments were performed preoperatively and at 6 weeks, 3 months, and 6 months postoperatively. Tertiary outcomes included the analysis of associations between the collected demographic, clinical, and surgical variables (e.g., Goutallier grade, tear size, comorbidities) and the primary outcomes (functional scores, ROM, re-tear).

2.6 Statistical analysis

Data were analyzed using SPSS software (version 26). Descriptive statistics were reported as mean ± standard deviation or frequency (percentage). The independent samples t-test and Chi-square test were used for between-group comparisons of continuous and categorical variables, respectively. The paired t-test or Wilcoxon test was used for within-group comparisons. A p-value of less than 0.05 was considered statistically significant. A multivariate logistic regression analysis was performed to control for potential confounders.

Subgroup analyses were performed based on tear size (<3 cm vs. ≥3 cm) and Goutallier grade (<2 vs. ≥2). To identify predictors of excellent outcome (UCLA score ≥33) and re-tear, univariate analysis followed by multivariate logistic regression models were constructed, including variables with p < 0.1 in the univariate screening (e.g., rehabilitation group, age, tear size, Goutallier grade, smoking, diabetes).

3.1 Baseline characteristics: comprehensive profile

No significant differences were observed between the EMG and DMG in all baseline demographic, clinical, and surgical characteristics (p > 0.05), confirming successful randomization (Table 1).

3.2 Primary outcomes at 6 months(Table 2)

•Pain (VAS): The EMG reported significantly lower pain scores compared to the DMG (1.7 ± 0.8 vs. 2.9 ± 1.1, p = 0.02).•Function (UCLA Score): The EMG achieved significantly higher UCLA scores (33.7 ± 2.1 vs. 29.4 ± 3.3, p = 0.01).•Range of Motion: The EMG demonstrated significantly greater ROM in forward flexion (165.8° ± 8.3 vs. 156.2° ± 5.2, p = 0.008), abduction (158.7° ± 4.4 vs. 152.3° ± 6.9, p = 0.04), external rotation (67.7° ± 4.2 vs. 63.3° ± 5.8, p = 0.03), and internal rotation (63.8° ± 3.1 vs. 59.4° ± 4.8, p = 0.04).•Structural Integrity (Re-tear): MRI assessment revealed no significant difference in the re-tear rate (Sugaya IV & V) between the EMG (3.3 %, n = 2) and the DMG (1.7 %, n = 1) (p = 0.48).

3.3 Secondary outcomes(Table 3)

•Patient Satisfaction: A significantly higher proportion of patients in the EMG reported “excellent” satisfaction (47 % vs. 33 %, p = 0.03).•Complications: The incidence of shoulder stiffness (defined as >20° deficit in active forward flexion/abduction or >15° deficit in external/internal rotation compared to the contralateral side) was significantly lower in the EMG (5 % vs. 15 %, p = 0.03). No significant differences were found in rates of superficial infection or persistent pain.•Return to Activities: Patients in the EMG returned to daily activities significantly earlier than those in the DMG (6.3 ± 1.4 weeks vs. 10.4 ± 2.7 weeks, p = 0.02).

Multivariate regression analysis confirmed that the early motion protocol was independently associated with better functional outcomes and a lower risk of stiffness, even after controlling for age, gender, surgical technique, and initial tear size.

3.4 Subgroup analysis: Surgical technique

Table 4 presents the primary functional outcome (UCLA score) stratified by the surgical approach, demonstrating the consistent benefit of early motion across techniques.

3.5 Analysis of Re-tear cases and high-risk factors

The re-tear rate was not significantly different between groups (EMG: 3.3 % vs. DMG: 1.7 %; p = 0.56). A detailed analysis of the three re-tear cases revealed a consistent profile (Table 5).

3.6 Predictors of excellent outcome and complications

Multivariate logistic regression identified the following independent predictors:•For Excellent Outcome (UCLA ≥33): Early Motion Protocol (OR = 4.2, 95 % CI 1.9–9.1; p = 0.001), Lower Pre-op VAS (OR = 1.8, 95 % CI 1.1–2.9; p = 0.02), and Goutallier Grade <2 (OR = 3.5, 95 % CI 1.5–8.0; p = 0.004).•For Postoperative Stiffness: Delayed Motion Protocol (OR = 3.8, 95 % CI 1.2–12.0; p = 0.02) and Diabetic Comorbidity (OR = 2.9, 95 % CI 1.0–8.4; p = 0.047).•For Re-tear: Due to the low event rate (n = 3), a robust model could not be fitted. Univariate analysis showed strong associations with Large Tear Size (p < 0.001) and Goutallier Grade ≥3 (p = 0.01).